> it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power
We currently make around 1 TW of photovoltaic cells per year, globally. The proposal here is to launch that much to space every 9 hours, complete with attached computers, continuously, from the moon.
edit: Also, this would capture a very trivial percentage of the Sun's power. A few trillionths per year.
We also shouldn't overlook the fact that the proposal entirely glosses over the implication of the alternative benefits we might realize if humanity achieved the incredible engineering and technical capacity necessary to make this version of space AI happen.
Think about it. Elon conjures up a vision of the future where we've managed to increase our solar cell manufacturing capacity by two whole orders of magnitude and have the space launch capability for all of it along with tons and tons of other stuff and the best he comes up with is...GPUs in orbit?
This is essentially the superhero gadget technology problem, where comic books and movies gloss over the the civilization changing implications of some technology the hero invents to punch bad guys harder. Don't get me wrong, the idea of orbiting data centers is kind of cool if we can pull it off. But being able to pull if off implies an ability to do a lot more interesting things. The problem is that this is both wildly overambitious and somehow incredibly myopic at the same time.
A lot of great inventions we now take for granted initially came with little motivation other than being able to kill each other more effectively. GPS, radar, jet engines, drones, super glue, microwaves, canned food, computers, even the internet. Contrary to the narrative of the internet being about sharing science, ARPANET was pushed by the DoD as a means of maintaining comms during nuclear war. It was then adopted by universities and research labs and started along the trajectory most are more familiar with.
The tale of computers is even more absurd. The first programmable, electric, and general-purpose digital computer was ENIAC. [1] It was built to... calculate artillery firing tables. I expect in the future that the idea of putting a bunch of solar into space to run GPUs for LLMs will probably seem, at the minimum - quaint, but that doesn't mean the story ends there.
That’s not the point of the person you are replying to. They are saying if we somehow come up with the tech that makes harnessing the sun a thing, the best we can still do is put a bunch of GPUs in space? It makes no sense.
oh, we'll sure find a way to weaponize that energy for example - just imagine all those panels simultaneously turning their reflective back in a way to form gigantic mirror to focus reflected solar energy on your enemy, be that enemy in space or on the Earth/Moon/Mars ground. Basically space-scale version of 'death ray scyscrapper' https://www.businessinsider.com/death-ray-skyscraper-is-wrea....
I think the Colossus[1] predated the ENIAC but is still in line with your general theme of doing stuff for the military. In this case it was used for cipher breaking, not firing calculations.
You could argue that it doesn't really count though because it was only turing complete in theory: "A Colossus computer was thus not a fully Turing complete machine. However, University of San Francisco professor Benjamin Wells has shown that if all ten Colossus machines made were rearranged in a specific cluster, then the entire set of computers could have simulated a universal Turing machine, and thus be Turing complete."
Not to go heads I win, tails you lose, but even if we go down this path - it's the same story because militaries are investing heavily in LLM stuff, both overtly and covertly. Outside of its obvious uses in modeling, data management, and other such things - there also seems to be a fairly widespread belief, among the powers that be, that if you just say the magic words to somebody, that you can make them believe anything. So hyper-scaling LLM potential has direct military application, same as Starlink and Starship.
This is a question that analysts don't even ask on earnings calls for companies with lowly earthbound datacenters full of the same GPUs.
The stock moves based on the same promise that's already unchecked without this new "in space" suffix:
We'll build datacenters using money we don't have yet, fill them with GPUs we haven't secured or even sourced, power them with infrastructure that can't be built in the promised time, and profit on their inference time over an ever-increasing (on paper) lifespan.
Not really. GPUs are stateless so your bounded lifetime regardless of how much you use them is the lifetime of the shitties capacitor on there (essentially). Modulo a design defect or manufacturing defect, I’d expect a usable lifetime of at least 10 years, well beyond the manufacturer’s desire to support the drivers for it (ie the sw should “fail” first).
A "fully and rapidly reusable" Starship would bring the cost of launch down orders of magnitude, perhaps to a level where it makes sense to send up satellites to repair/refuel other satellites.
Nope, it's 100% about building the stock valuation of SpaceX for an IPO in the face of significant risk from a cold war its CEO started on X with the U.S. federal government and increasing competition from Blue Origin, Quinfan and Guowang. DOD will play Bedrock vs Grok until there is feature parity and then make a decision not based on the features.
Disclaimer: Not an Elon hater, but far from a sycophant, similar to how I felt about Steve Jobs for 40+ years.
Exactly, this is about attaching the AI hype bubble to all of his dealings before he offloads with an IPO (that still leaves him with 75% of the stock).
It's always better to generate electricity on the ground than attempt to beam it to the ground from space. The efficiency loss of beamed power is huge.
The efficiency loss of nighttime is approximately 100% if we’re talking about solar energy. At least at a most basic level, it’s not totally absurd to stick some kind of power beaming contraption in space where it is mostly not shadowed by the Earth and beam power to a ground station.
I concur it’s not necessarily totally absurd — but when you consider that such contraptions require large — very large! — receiving arrays to be built on the ground, it’s hard to avoid concluding that building gigantic photovoltaic arrays in, say Arizona (for the US) along with batteries for overnight buffering and transmission lines would still be massively more efficient.
Is that more or less absurd than making deals with our neighbours to share their electricity? Build some solar farms around the planet and then distribute it over wire.
I honestly don't know the answer. I know there's some efficiency loss running over long wires too but I don't know what's more realistic.
We have these things called batteries, you charge them during the day, and drain them at night.
A solar+battery setup is already cheaper than a new gas plant. Beaming power from space is absolutely asinine, quite frankly. The losses are absurd, the sun already does it 24/7, and we know how to make wires and batteries to shuffle the sun's power around however we need to. Why on earth would we involve satellites?
Right up to the radiation limit and then you'll either have to throttle your precious GPUs or you'll be melting your satellite or at least the guts of it. You're looking at an absolutely massive radiator here, many times larger than the solar panels that collect the energy to begin with.
If we (as in "civilization") were able to produce that many solar panels, we should cover all the deserts with them. It will also shift the local climate balance towards a more habitable ecosystem, enabling first vegetation and then slowly growing the rest of the food chain.
> But being able to pull if off implies an ability to do a lot more interesting things.
Those interesting things won't pump up the perceived value of Musk companies to stratospheric levels - or dare I say - to the moon. He needs the public to believe that to earn the trillion-dollar package from the Tesla-Twitter-SpaceX conglomerate, even if the latter turns out to be the only profitable arm of the conglomerate.
You really can't grasp that GPUs scaled at this level is the most ambitious thing possible? That it will be the foundation of unfathomable technological innovation?
Honestly, there's not a lot else I can think of if your goal is find some practical and profitable way to take advantage of relatively cheap access to near-Earth space. Communication is a big one, but Starlink is already doing that.
One of the things space has going for it is abundant cheap energy in the form of solar power. What can you do with megawatts of power in space though? What would you do with it? People have thought about beaming it back to Earth, but you'd take a big efficiency hit.
AI training needs lots of power, and it's not latency sensitive. That makes it a good candidate for space-based compute.
I'm willing to believe it's the best low-hanging fruit at the moment. You don't need any major technological advances to build a proof-of-concept. Whether it's possible for this to work well enough that it's actually cheaper than an equivalent terrestrial datacenter now or in the near future is something I can't answer.
We have radiators on the ISS. Even if you kept the terrible performance of those ancient radiator designs (regularly exposed to sunlight, simplistic ammonia coolant, low temperature) you could just make them bigger and radiate the needed energy. Yes it would require a bit of engineering but to call it an "unsolved problem" is just exaggerating.
It's a solved problem. The physics is simply such that it's really inefficient.
> ... we'd need a system 12.5 times bigger, i.e., roughly 531 square metres, or about 2.6 times the size of the relevant solar array. This is now going to be a very large satellite, dwarfing the ISS in area, all for the equivalent of three standard server racks on Earth.
The gist of it is that about 99% of cooling on earth works by cold air molecules (or water) bumping into hot ones, and transferring heat. There's no air in space, so you need a radiator 99x larger than you would down here. That adds up real fast.
That’s the secret plan - cover LEO with solar cells and radiators, limiting sunlight on the ground, rendering ground base solar ineffective, cool earth and create more demand for heating; then sell expensive space electricity at a huge premium. Genius!
Bezos has been pushing manufacturing-in-space for a long time, as a ideal candidate for what to do in space that you might prefer to not do on Earth. Robotics, AI automation, manufacturing - combo it in space, let the robots manufacture for us in space. Abundant energy, low concerns about most forms of pollution. We'll need to dramatically improve our ability to transit mass to and from cheaply first of course (we're obviously talking many decades into the future).
That is a fun thought experiment, as we wouldn't want to manufacture too far away from earth we may still be within the earth's atmosphere. I wonder what effect dumping greenhouse gases into the very upper levels of the atmosphere would have in comparison to doing it lower down. My assumption is it would eventually sink to a lower density layer, having more or less the same impact.
When Bezos first mentioned drone delivery, many intelligent, serious people laughed at it and accused of Bezos running out of ideas as Amazon was stagnant
> Bezos has been pushing manufacturing-in-space for a long time, as a ideal candidate for what to do in space that you might prefer to not do on Earth. Robotics, AI automation, manufacturing - combo it in space, let the robots manufacture for us in space.
LOL, this seems so far off from the reality of what manufacturing looks like in reality.
- sending raw materials up there
- service technicians are necessary ALL THE TIME, in fully automated production lines
- sending stuff back down
Maybe I lack vision, but data centers in space is a 1000x times better idea and that is already a terrible idea.
The show For All Mankind kind-of hinted at how the labor problem would be solved: recruit like the military and promise huge bonuses that will probably not be realized because space is risky business
I think it makes more sense if you invert the manufacturing cycle.
Automated asteroid mining, and asteroid harvesting, are potential areas where we have strong tech, a reasonable pure automation story, and huge financial upsides. Trillion dollar asteroids... If we’re sourcing metals out there, and producing for orbital operations or interplanetary shenanigans, the need for computing and automation up there emerges.
And I imagine for the billionaire investor class now is the window to make those kinds of plays. A whole set of galactic robber barons is gonna be crowned, and orbital automation is critical to deciding who that is.
But everyone is crazy about GPU’s right now. Why not ride that wave for extra investment? All the benefits transfer to all the other things we can do with it.
We also shouldn't overlook the benefits we might realize if humanity achieved the incredible engineering and technical capacity necessary to make this version of porcine flight happen.
That's not a new problem that no one has dealt with before. The ISS for instance has its External Active Thermal Control System (EACTS).
It's not so much a matter of whether it's an unsolvable problem but more like, how expensive is it to solve this problem, what are its limitations, and does the project still makes economic sense once you factor all that in?
It's worth noting that the EACTS can at maximum dissipate 70kW of waste heat. And EEACTS (the original heat exchange system) can only dissipate another 14kW.
That is together less than a single AI inference rack.
And to achieve that the EACTS needs 6 radiator ORUs each spanning 23 meters by 11 meters and with a mass of 1100 kg. So that's 1500 square meters and 6 and a half metric tons before you factor in any of the actual refrigerant, pumps, support beams, valve assemblies, rotary joints, or cold side heat exchangers all of which will probably together double the mass you need to put in orbit.
There is no situation where that makes sense.
-----------
Manufacturing in space makes sense (all kinds of techniques are theoretically easier in zero G and hard vacuum).
Mining asteroids, etc makes sense.
Datacenters in space for people on earth? That's just stupid.
I get that vacuum is a really good insulator, which is why we use it to insulate our drinks bottles. So disposing of the heat is a problem.
Can't we use it, though? Like, I dunno, to take a really stupid example: boil water and run a turbine with the waste heat? Convert some of it back to electricity?
What do you do with the steam afterwards? If you eject it, you have to bring lots of it with your spacecraft, and that costs serious money. If you let it condensate to get water again, all you did is moving some heat inside the spacecraft, almost certainly creating even more heat when doing that.
It's a good question, but in a closed system (like you have in space) the heat from the turbine loop has to go somewhere in order to make it useful. Let's say you have a coolant loop for the gpus (maybe glycol). You take the hot glycol, run it through your heat exchanger and heat up your cool, pressurized ammonia. The ammonia gets hot (and now the glycol is cool, send it back). You then take the ammonia and send it through the turbine and it evaporates as it expands and loses pressure to spin the turbine. But now what? You have warm, vaporized, low pressure ammonia, and now you need to cool it down to start over. Once it's cool you can pressurize it again so you can heat it up to use again, but you have to cool it, and that's the crux of the issue.
The problem is essentially that everything you do releases waste heat, so you either reject it, or everything continues to heat up until something breaks. Developing useful work from that heat only helps if it helps reject it, but it's more efficient to reject it immediately.
A better, more direct way to think about this might be to look at the Seebeck effect. If you have a giant radiator, you could put a Peltier module between it and you GPU cooling loop and generate a little electricity, but that would necessarily also create some waste heat, so you're better off cooling the GPU directly.
The ISS consumes roughly 90kW. That’s about *one* modern AI/ML server rack. To do that they need 1000 m^2 of radiator panels (EACTS). So that’s the math: every rack needs another square kilometer of stuff put into orbit. Doesn’t make sense to me.
And what happens every time a rack (or node) fails? Does someone go out and try to fix it? Do we just "deorbit" it? How many tons per second of crap would we be burning in the upper atmosphere now? What are the consequences of that?
How do the racks (or nodes) talk to eachother? Radios? Lasers?
What about the Kessler Syndrome?
Not a rocket scientist but 100% agree this sounds like a dead end.
Communication is a well-understood problem, and SpaceX already has Starlink. They might need pretty high bandwidth, but that's not necessarily much of a problem in space. Latency could be a problem, except that AI training isn't the sort of problem where you care about latency.
I'd be curious where exactly they plan to put these datacenters... In low Earth orbit they would eventually reenter, which makes them a pollution source and you'd have no solar power half the time.
Parking them at the Earth-Sun L1 point would be better for solar power, but it would be more expensive to get stuff there.
Seasons mess that up unless you're burning fuel to make minor plane changes every day. Otherwise you have an equinox where your plane faces the sun (equivalent to an equatorial orbit) and a solstice where your plane is parallel to the sun (the ideal case).
It makes sense to target a higher operating temperature, like 375K. At some point, the energy budget would reach an equilibrium. The Earth constantly absorbs solar energy and also dissipates the heat only by radiative cooling. But the equilibrium temperature of the Earth is still kind of cool.
I guess the trick lies in the operating temperature and the geometry of the satellites.
Asking for a friend (who sucks at thermodynamics:) could you use a heat pump to cool down the cold end more and heat up the hot end much higher? Heat radiation works better the higher the temperature?
The distinction is that what they are doing for Webb is trying to dissipate small amounts of heat that would warm up sensors past cryogenic temperatures.
Like on the order of tens or hundreds of watts but -100C.
Dissipating heat for an AI datacenter is a different game. A single AI inference or training rack is going to be putting out somewhere around 100kW of waste heat. Temps don't have to be cryogenic but it's the difference between chiselling a marble or jade statue and excavating a quarry.
That's a solution for minuscule amounts of heat that nevertheless disturb extremely sensitive scientific experiments. Using gold, no less. This does not scale to a crapton of GPU waste heat.
Anything is possible here, it's just there's no goddamn reason to do any of this. You're giving up the easiest means of cooling for no benefit and you add other big downsides.
It's scifi nonsense for no purpose other than to sound cool.
Getting better at creating and erecting solar panels & AI datacenters on earth is all well and good, but it doesn't advance SpaceX or humanity very much. At lot of the bottlenecks there are around moving physical mass and paperwork.
Whereas combining SpaceX & xAI together means the margins for AI are used to force the economies of scale which drives the manufacturing efficiencies needed to drive down launch etc.
Which opens up new markets like Mars etc.
It is also pushing their competitive advantage. It leaves a massive moat which makes it very hard for competitors. If xAI ends up with a lower cost of capital (big if - like Amazon this might take 20 years horizon to realize) but it would give them a massive moat to be vertically integrated. OpenAI and others would be priced out.
If xAI wants to double AI capacity then it's a purely an automation of manufacturing problem which plays to Elons strengths (Tesla & automation). For anyone on earth doubling capacity means working with electricity restrictions, licensing, bureaucracy, etc. For example all turbines needed for electricity plants are sold years in advance. You can't get a new thermal plant built & online within 5 years even if you had infinite money as turbines are highly complex and just not available.
Hmm, Elon really did run that flywheel pretty well. He built the Roadster to drum up some cash and excitement so he could develop the Model S, then he used that success to do the Model X, and then he expanded capacity to develop the 3 and Y, and he reinvested the profits to develop the Model 2, finally bringing EVs to the masses, displacing ICEs everywhere, and becoming the undisputed leader of both EV and battery manufacturering.
Oh wait, that didn’t actually happen, because he got distracted or something? He doesn’t really have battery capacity worth writing home about, the Chinese are surpassing Tesla in EV manufacturing, and Waymo is far ahead in self-driving.
The amazing space computation cost reduction process sounds rather more challenging than the Model 2, and I’m not sure why anyone should bet on Elon pulling it off.
Tesla invested into the first Lotus roadster - and put that cash into the S then the X. Used that cash to build the worlds largest factories and make the 3 & Y which sold at enormous volumes - so large in fact that the S & X are now tiny single percentages of sales which is why Tesla is stopping manufacturing them now.
Tesla is one of the very few vehicle manufactures which makes a profit manufacturing vehicles. Tesla throws off cash which allows the flywheel to keep spinning.
Tesla is now operating fully autonomous rides. They've constantly proved their naysayers wrong at every turn in time. What the Chinese are doing in battery tech is irrelevant to US vehicles as they will never be allowed to sell in the US which is Teslas largest market.
The model 2 has the possibility of being profitable at insanely low purchase price which has the potential to completely disrupt the economics of US sales in such a way that legacy auto could well be bankrupt in 5-10 years. Who will be making Waymo's vehicles then?
Hey remember that time someone had their Tesla running down the highway and the superior self-driving capability failed to see an 18 wheeler that crossed the road and the person was decapitated and there are videos of that complete with blood spray?
There is nothing we need on Mars other than science. It's not a market because there isn't money to be made outside of what is required to do whatever economically useless but scientifically valuable efforts we can convince people to fund.
We can't build an independent colony we can't live there any time soon. Arguably it may never make sense to live there.
1. Mankind never systematically lived in caves; that's just where remains and rock paintings are more likely to have survived.
2. Farming didn't evolve from a vision of "let's stay in one place, so let's find a way to do it"; it evolved from the gradual application of accumulated practical knowledge under real constraints until eventually it was possible to stay in one place. If Paleoelon had somehow convinced early humanity to abandon hunter-gathering and settle into a sedentary life because he had a vision for new markets around farming it would have led to the earliest famine.
All right, so how is it that all you geniuses out here are totally right about this, but all the dullards at SpaceX and XAI, who have accomplished nothing compared to you lot, are somehow wrong about what they do every day?
I know being right without responsibility feels amazing but results are a brutal filter.
I once had a job mopping floors and was quite successful at it, even if I say so myself. Based on my experience, do you think it is reasonable for me to claim that I will eventually develop techniques for cleaning the oceans of all plastic waste? Folks are criticizing the pie in the sky claims, not that they can do anything at all.
Seems a bit of both. But no disparagement to your floor mopping (as I once was a dishwasher in a commercial kitchen myself), but there's a big gap between cleaning a floor, or a dish, and creating frontier models and spaceships.
That said: I think solar is niche, and a moon-shot for how they want it. Nuclear is the future of reliable energy for human civilization.
I think the K-scale is the wrong metric. I don't think we should be trying to take all the sun's energy as a goal (don't blot out the sun! don't hide it in a bushel!), or as a civilizational utiltiy - I'm sure better power supplies will come along.
Data centers ultimately need to provide power and remove heat. Solar might be a little easier for power in space, maybe, but heat is an absolute no-go, stop, this will never ever work. You can't engineer your way out of the fact that space is a vacuum.
There's no reason to think the brilliant minds at SpaceX are supportive of focusing their mission in any manner-what-so-ever on datacenters in space. You can't call on their genius as the supportive argument accordingly.
I disagree, I think the idea of a cabal of reactionary comrades inside SpaceX is activist fantasy. I think SpaceX only does what it does with full committment of its people: mind, body, spirit.
I think there's a scenario where that's true: one where the head of your company is collaborative and deferential to expertise.
There's another scenario, though: one where the head of your company is a bull in a China shop, whose successes have come almost exclusively through a Barnum-esque scheme of cascading bravado and marketing genius without much expertise, but a marvelous ability to sell any idea purely via unearned gravitas.
The former is less sexy: I've compiled loads of talented people, and we're going to solve very hard problems, even some that seem impossible.
The latter is very sexy: I'm a genius and we're going to accomplish the impossible in one year via sheer force of my grand will. And even if it doesn't actually happen, I'll sell you on the next vision.
It seems like you’re ascribing to Elon some kind of magic, where you feel he’s breaking the rules of what should be allowed in order to achieve success. Is it impossible you simply don’t understand how what he does works?
Microsoft did do the experiment (Project Natick) where they had "datacenters" in pods under the sea with really good results. The idea was simply to ship enough extra capacity, but due to the environment, the failure rates where 1/8th of normal.
Still, dropping a pod into the sea makes more sense than launching it into space. At least cooling, power, connectivity and eventual maintenance is simpler.
The whole thing makes no sense and is seems like it's just Musk doing financial manipulation again.
Haha, hard pass on the job. I prefer my oxygen at 1 atm.
I'm not a data center technician myself, but I have deep respect for those folks and the complexity they manage. It's quite surprising the market still buys Musk's claims day after day.
There are a class of people who may seem smart until they start talking about a subject you know about. Hank Green is a great example of this.
For many on HN, Elon buying Twitter was a wake up call because he suddenly started talking about software and servers and data centers and reliability and a ton of people with experience with those things were like "oh... this guy's an idiot".
Data centers in space are exactly like this. Your comment (correctly) alludes to this.
Companies like Google, Meta, Amazon and Microsoft all have so many servers that parts are failing constantly. They fail so often on large scales that it's expected things like a hard drive will fail while a single job might be running.
So all of these companies build systems to detect failures, disable running on that node until it's fixed, alerting someone to what the problem is and then bringing the node back online once the problem it's addressed. Everything will fail. Hard drives, RAM, CPUs, GPUs, SSDs, power supplies, fans, NICs, cables, etc.
So all data centers will have a number of technicians who are constantly fixing problems. IIRC Google's ratio tended to be about 10,000 servers per technician. Good technicians could handle higher ratios. When a node goes offline it's not clear why. Techs would take known good parts and basically replacce all of them and then figure out what the problem is later, dispose of any bad parts and put tested good parts into the pool of known good parts for a later incident.
Data centers in space lose all of this ability. So if you have a large number of orbital servers, they're going to be failing constantly with no ability to fix them. You can really only deorbit them and replace them and that gets real expensive.
Electronics and chips on satellites also aren't consumer grade. They're not even enterprise grade. They're orders of magnitude more reliable than that because they have to deal with error correction terrestial components don't due to cosmic rays and the solar wind. That's why they're a fraction of the power of something you can buy from Amazon but they cost 1000x as much. Because they need to last years and not fail, something no home computer or data center server has to deal with.
Put it this way, a hardened satellite or probe CPU is like paying $1 million for a Raspberry Pi.
And anybody who has dealt with data centers knows this.
Great comment on hardware and maintenance costs, and in comparison Elon wrote "My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space."
It's a pity this reads like the entire acquisition of xAi is based on "Elon's napkin math" (maybe he checked it with Grok)
The deal they made values xAI at $230 Billion. It’s a made up number, with no trustworthy financial justification to back it up. It is set to provide a certain return to xAI’s investors (the valuation decides the amount you get per share), who in turn are bailing out the earlier acquisition of X (Twitter). All of this is basically a shell game where Elon is using one company to bail out another. It’s a way of reducing the risk of new ventures by spreading them out between his companies. It’s also really bad for SpaceX employees and investors, who are basically subsidizing other companies.
The thing is, everyone knows Elon is not a real CEO of any of these companies. There isn’t enough time to even be the CEO of one company and a parent. This guy has 10 companies and 10 children. He’s just holding the position and preventing others from being in that position, so he can enact changes like this. And his boards are all stacked with family members, close friends, and sycophants who won’t oppose his agenda.
Most of the investors don’t even have a choice. Nor do all the other shareholders like employees. And the boards of Musk companies are stacked with his yes men.
He's bailing out one of his failing ventures with one of his so far successful ones. The BS napkin math isn't the reason he's doing it. It's the excuse for doing it.
Thanks for putting words to that; the paragraph which most stuck out to me as outlandish is (emphasis mine):
The basic math is that launching a million tons per year of satellites generating 100 kW of compute power per ton would add 100 gigawatts of AI compute capacity annually, *with no ongoing operational or maintenance needs*.
I'm deeply disillusioned to arrive at this conclusion but the Occam's Razor in me feels this whole acquisition is more likely a play to increase the perceptual value of SpaceX before a planned IPO.
First of all Twitter had basically no downtime since he bought it, so all the 'internet experts' posting their thoughts were completely dead wrong. If anything Twitter was far more reliable than Microsoft has been these past few years.
You are assuming things need to run the same way in space, for instance you mentioned fans, you won't have any in space. You also won't have any air, dust, static, or any moving parts.
You are assuming the costs to launch to orbit are high, when the entire point of Spacex's latest ship is to bring the cost to launch so low that it is cheaper per ton than an airplane flight.
Maintenance would be nice but you are saying this like Elon Musk's company doesn't already manage the most powerful datacenters on the planet.
You have no clue what you are talking about regarding cosmic rays and solar wind, these will literally be solar powered and behind panels and shielding 100% of the time.
I went looking through your comments. 75% of them (and probably 90% in the lasst 2 years) were Elon related. Tesla, SpaceX, Grok, Twitter, DOGE, etc. Quite a lot of comments for 101 karma if I'm being real.
Why do you feel this kneejerk reaction to defend Elon and his companies? You'll never be him. He doesn't care about you. He'd use you for reactor shielding for an uptick in Tesla share price without a second's hesitation. This is cultish behavior.
Do you have any idea who you're defending? I'll give you just one example. A right-wing influencer named Dom Lucre uploaded CSAM to Twitter, a video. But he didn't just upload it. He watermarked it first so had it on his computer and then postporcessed it. It was I believe up for days. This was apparently a video so bad that mere possession should land you in prison. And the fact that the FBI didn't arrest him basically tells you he'd an FBI asset. After taking days to ban him, Elon personally intervened to unban him. Why? Because reasons.
And this is the same man who it's becoming clear was deeply linked with Jeffrey Epstein, as was his brother [1].
Bringing this back to the original point: this is why Twitter lost 80% of its value after Elon acquired it. Advertisers fled because it became a shithole for CSAM and Nazis.
As for "basically no downtime" that's hilarious. I even found you commenting the classic anecdote "it was fine for me" (paraphrased) on one such incident when Twitter DDOSed itself [2].
Your cultish devotion here is pretty obvious eg [3]. I'm genuinely asking: what do you get out of all this?
Maintaining modern accelerators requires frequent hands-on intervention -- replacing hardware, reseating chips, and checking cable integrity.
Because these platforms are experimental and rapidly evolving, they aren't 'space-ready.' Space-grade hardware must be 'rad-hardened' and proven over years of testing.
By the time an accelerator is reliable enough for orbit, it’s several generations obsolete, making it nearly impossible to compete or turn a profit against ground-based clusters.
On the other hand, Tesla vehicles have similar hardware built into them, and don't require such hands-on intervention. (And that's the hardware that will be going up.)
Car-grade inference hardware is fundamentally different from data center-grade inference hardware, let alone the specialized, interconnected hardware used for training (like NVLink or complex optical fabrics). These are different beasts in terms of power density, thermal stress, and signaling sensitivity.
Beyond that, we don't actually know the failure rate of the Tesla fleet. I’ve never had a personal computer fail from use in my life, but that’s just anecdotal and holds no weight against the law of large numbers. When you operate at the scale of a massive cluster, "one-in-a-million" failures become a daily statistical certainty.
Claiming that because you don't personally see cars failing on the side of the road means they require zero intervention actually proves my original point: people who haven't managed data center reliability underestimate the sheer volume of "rare" failures that occur at scale.
Do they need to be maintained? If one compute node breaks, you just turn it off and don't worry about it. You just assume you'll have some amount of unrecoverable errors and build that into the cost/benefit analysis. As long as failures are in line with projections, it's baked in as a cost of doing business.
The idea itself may be sound, though that's unrelated to the question of whether Elon Musk can be relied on to be honest with investors about what their real failure projections and cost estimates are and whether it actually makes financial sense to do this now or in the near future.
AI clusters are heavily interconnected, the blast radius for single component failure is much larger than running single nodes -- you would fragment it beyond recovery to be able to use it meaningfully.
I can't get in detail about real numbers but it's not doable with current hardware by a large margin.
Yeah, I remember people saying that about making 1m model 3s per year, landing rockets, getting 10k+ satellite privately into orbit, and getting millions of subscribers using internet via those satellites.
Maybe just maybe the guy does actually get things done, and if you didn't hate him you'd see that?
(yes, there are some things he hasn't gotten done. That doesn't take away from what he has gotten done)
Context missing. This is in reference to a vision the (distant?) future where the satellites are manufactured in factories on the Moon and sent into space with mass drivers.
Full paragraph quote comes from:
> While launching AI satellites from Earth is the immediate focus, Starship’s capabilities will also enable operations on other worlds. Thanks to advancements like in-space propellant transfer, Starship will be capable of landing massive amounts of cargo on the Moon. Once there, it will be possible to establish a permanent presence for scientific and manufacturing pursuits. Factories on the Moon can take advantage of lunar resources to manufacture satellites and deploy them further into space. By using an electromagnetic mass driver and lunar manufacturing, it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power.
>
> This is in reference to a vision the (distant?) future where the satellites are manufactured in factories on the Moon and sent into space with mass drivers.
In the meantime, how about affordable insulin for everybody?
Why is it cheaper to ship all of the materials to space, then to the moon for assembly (which also includes shipping all of the people and supplies to keep them alive), then back into space vs just…
It's not like satellites need anything like computer chips, which are finicky things to build that require parts with a sole supplier on the entire planet.
You can make propellant on the Moon (aluminum based solid fuels), and the energy to get into orbit or into deep space is far, far less that from Earth’s surface.
Why would satellites be manufactured on the moon? There's nothing on the moon. The raw materials would have to be ferried over first. What would be the point?
It would appeal to naive technofetishists, the same crowd of investors enamored by many of Elon's other impossible schemes.
The moon mfg makes significantly more sense than the hilarious plan to establish a permanent Mars base in the next 50 years, but that's not saying much.
> Why would satellites be manufactured on the moon? There's nothing on the moon. The raw materials would have to be ferried over first. What would be the point?
From lunar regolith you would extract: oxygen, iron, aluminum, titanium, silicon, calcium, and magnesium.
From the poles you can get fuel (water ice -> water + hydrogen + oxygen).
The real constraint is not materials, but rather power generation, automation reliability, and initial capital investment.
So you have to shuttle machines, energy systems, and electronics.
The moon can supply mass, oxygen, fuel, and structure.
Satellites that would benefit most are: huge comms platforms, space-based power satellites, large radar arrays, deep-space telescopes, etc.
Power would almost certainly mostly come from solar panels. The SpaceX-xAI press release mentions using mass drivers which are electrically powered. Could make Hydrogen-Oxygen rocket fuel but not needed in Moon's lower gravity/thin atmosphere.
It's solvents, lubricants, cooling, and all the other boring industrial components and feedstocks that people seem to forget exist. Just because raw materials exist in lunar regolith doesn't mean much if you can't actually smelt and refine it into useful forms.
> We currently make around 1 TW of photovoltaic cells per year, globally.
Doubling every three years; at that rate it would take about 30 years for 1TW to become 1000TW. Whether on not the trend continues largely depends on demand, but as of right now humanity seems to have an insatiable demand for power.
We’re not going to use 100% of our solar panel manufacturing capacity to power space data centers, specifically because everyone else on the ground is so power-hungry. If we’re being generous, it could maybe top out at 1%, which adds another ~20 years to your timeline for a total of 50. I think it’s safe to say this part is bunk (along with everything else about this plan which is also bunk).
I wonder what the plan is to recycle those. Without a plan to safely bring back all this hardware and recycling it, we'll deplete earth from it's mineral. The matter used to build things on earth stays within earth's ecosystem.
Moving matter out continusously at industrial scale with no plan to bring 100% of it back in the ecosystem other than burning it seems quite unsustainable and irresponsable.
Whilst I agree that this glosses over a huge number of technical obstacles, space based solar power could scale more easily than that on earth. Lack of variable weather and gravity means rather than using photovoltaic cells, you can just set up paper thin huge mirrors to focus light and generate steam.
Caveat: my understanding of this largely comes from the book The High Frontier, which is really old and probably inaccurate. I can't think of a reason why this particular point would be wrong though.
Earth does have plenty of sand and iron. Literally all you have to do is grow the sand into a crystal, slice it up, etch some patterns onto it, then add some metal.
Making only 1TW of pv cells per year is a skill issue.
The quoted "1 TW of photovoltaic cells per year, globally" is the peak output, not the average output. They're only about 20% higher peak output in space… well, if you can keep them cool at least.
The 1TW is the rated peak power output. It's essentially the same in space. The thing that changes is the average fraction of this sustained over time (due to day/night/seasons/atmosphere, or the lack of all of the above).
It's still the same 1TW theoretical peak in space, it's just that you can actually use close to that full capacity all the time, whereas on earth you'd need to over-provision substantially and add storage, so 1TW of panels can only drive perhaps a few hundred GW of average load.
The dominant factor is "balance of system" aka soft costs, which are well over 50%.[0]
Orbit gets you the advantage of 1/5th the PV and no large daily smoothing battery, but also no on-site installation cost, no grid interconnect fees, no custom engineering drawings, no environmental permitting fees, no grid of concrete footers, no heavy steel frames to resist wind and snow loads. The "on-site installation" is just the panels unfolding, and during launch they're compact so the support structure can be relatively lightweight.
When you cost building the datacenter alone, it's cheaper on earth. When you cost building the solar + batteries + datacenter, it (can be) cheaper in space, if you build it right and have cheap orbital launch.
I do say it's predicated on cheap orbital launch. Clearly they expect Starship to deliver, and they're "skating to where the puck will be" on overall system cost per unit of compute.
But yeah, I didn't include that delivering all that stuff by truck (including all the personnel) to a terrestrial PV site isn't free either.
Yeah, soft costs like permitting and inspections are supposedly the main reason US residential solar costs $3/watt while Australian residential solar costs $1/watt. It was definitely the worst and least efficient part of our solar install, everything else was pretty straightforward. Also, running a pretty sizable array at our house, the seasonal variation is huge, and seasonal battery storage isn’t really a thing.
Besides making PV much more consistent, the main thing this seems to avoid is just the red tape around developing at huge scale, and basically being totally sovereign, which seems like it might be more important as tensions around this stuff ramp up. There’s clearly a backlash brewing against terrestrial data centers driving up utility bills, at least on the East Coast of the US.
The more I think about it, the more this seems like maybe not a terrible idea.
Do you imagine there'd be less red tape involved in launching multiple rockets per day carrying heavy payloads?
Like this argument just gets absurd: you're claiming building a data center on earth will be harder from a permitting perspective than FAA flight approval for multiple heavy lift rocket launch and landing cycles.
Mining companies routinely open and close enormous surface area mines all over the world and manage permitting for that just fine.
There's plenty of land no one will care if your build anything on, and being remote with maybe poor access roads is still going to be enormously cheaper then launching a state of the art heavy lift rocket which doesn't actually exist yet.
This is really underselling it tbh. Any land that's growing corn in a developed country is likely top 1% of land on earth. Half of the earth is desert and tundra. Which is still incredibly easier to work with than space because you can ship there with a pickup very cheaply. Maybe when nevada and central australia are wall-to-wall solar panels we can check back on space.
The Technology Connections Youtube channel recently did a great video arguing pretty convincingly that the land used to grow corn for cars would be vastly more efficiently used from an energy perspective if we covered it with solar panels.
You just have to remember, most of these people live in high density regions and have little comprehension about how much surface area humanity truly occupies... And that isn't even accounting for offshore constructs.
Realizing the impracticality of it (and that such approaches often collapse under the infeasibility of it) ... wouldn't it be better to... say... cover the Sahara in solar panels instead? That's gotta be cheaper than shipping them into space.
From an engineering perspective, with today’s costs, yes. But don’t forget the political complications of dealing with all those countries that own the Sahara, that’s going to come at it’s own cost.
I think this is all ridiculous, to be clear, but re: this problem couldn't the radiators in theory be oriented so that they vent in opposite directions and cancel out any thrust that would be generated?
Solar modules you can buy for your house usually have quoted power ratings at "max STC" or Standard Testing Conditions, which are based on insolation on Earth's surface.
I think it's because at this scale a significant limit becomes the global production capacity for solar cells, and SpaceX is in the business of cheaper satellites and launch.
You don't even need a particularly large scale, it's efficient resource utilization.
Humanity has a finite (and too small) capacity for building solar panels. AI requires lots of power already. So the question is, do you want AI to consume X (where X is a pretty big chunk of the pie), or five times X, from that total supply?
Using less PV is great, but only if the total cost ends up cheaper than installing 5X the capacity as terrestrial PV farms, along with daily smoothing batteries.
SpaceX is only skating to where they predict the cost puck will be.
Satellites can adjust attitude so that the panels are always normal to the incident rays for maximum energy capture. And no weather/dust.
You also don't usually use the same exact kind of panels as terrestrial solar farms. Since you are going to space, you spend the extra money to get the highest possible efficiency in terms of W/kg. Terrestrial usually optimizes for W/$ nameplate capacity LCOE, which also includes installation and other costs.
For one or a few-off expensive satellites that are intended to last 10-20 years, then yes. But in this case the satellites will be more disposable and the game plan is to launch tons of them at the lowest cost per satellite and let the sheer numbers take care of reliability concerns.
It is similar to the biological tradeoff of having a few offspring and investing heavily in their safety and growth vs having thousands off offspring and investing nothing in their safety and growth.
> And then there’s that pesky night time and those annoying seasons.
The two options there are cluttering up the dawn dusk polar orbit more or going to high earth orbit so that you stay out of the shadow of the earth... and geostationary orbits are also in rather high demand.
I grew up on a rural farm in California with a dial-up connection that significantly hampered my ability to participate in the internet as a teenager. I got Starlink installed at my parents' house about five years ago, and it's resulted in me being able to spend considerably more time at home.
Even with their cheapest home plan, we're getting like 100 Mbps down and maybe 20 to 50 up. So it's just not true at all that you would have connections that are a megabit or two per second.
That's not what I'm suggesting. The post says "deep space". If you're going to try to harvest even a tiny percentage of the sun's energy, you're not doing that in Earth's orbit. The comparison is a webcam feed from Mars.
That's pretty much a solved problem. We've had geostationary constellations for TV broadcast at hundreds of megabytes for decades now, and lasers for sat-to-sat comms seems to be making decent progress as well.
> it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power
3. vLLMa exist and take video and images as input.
4. When a new model checkpoint needs to go up, are we supposed to wait months for it to transfer?
5. A one million token context window is ~4MB. That's a few milliseconds terrestrially. Assuming zero packet loss, that's many seconds
6. You're not using TCP for this because the round trip time is so high. So you can't cancel any jobs if a user disconnects.
7. How do you scale this? How many megabits has anyone actually ever successfully sent per second over the distances in question? We literally don't know how to get a data center worth of throughput to something not in our orbit, let alone more than double digit megabits per second.
The intractable problem is heat dissipation. There is to little matter in space to absorb excess heat. You'd need thermal fins bigger than the solar cells. The satellite's mass would be dominated by the solar panels and heat fins such that maybe 1% of the mass would be usable compute. It would be 1000x easier to leave them on the moon and dissipate into the ground and 100000x easier to just keep making them on earth.
and, of course and inter-satellite comms and earth base station links to get the data up and down. Starlink is one thing at just above LEO a few hundred km and 20km apart, but spreading these around 10s of thousands of km and thosands of km apart is another thing
We can make ten or hundred times the number of solar cells we make right now, we just don't have a reason to. The technology is fairly ancient unless you want to compete on efficiency, and the raw materials abundant.
Do we really need to say (on HN especially) that time-to-market does matter?
Not just for startups either. If you ramp up the Polio vaccine in 1 year vs 10 years, that has a big impact on human wellbeing. The two scenarios are not equivalent outcomes, even though it still happens "eventually."
The bottleneck is deploying solar physically, not making the cells.
We have increased the manufacturing of pretty much every piece of technology you see in front you by 200x at some point in history. Often in a matter of years.
I agree that part of the bottleneck is deploying solar physically. China is the best in the world in deploying solar panels. They are only managing linear increases in their solar capacity, year over year.
You're not considering some important multipliers. In space you're already getting a substantial immediate boost due to greater solar irradiance - no atmosphere or anything getting in the way of those juicy photons. You can also get 24 hour coverage in space. And finally they mention "deep space" - it's unclear what that means but solar irradiance increases on an inverse square law - get half way to the sun and you're getting another 4x boost in power. I'm sure there's other factors I'm not considering as well - space and solar just go quite well together.
This is all based on bad math. The people proposing these things don't even have proper scientific and mathematical training to determine what is achievable.
>In an interview with Robert Wright in 2003, Dyson referred to his paper on the search for Dyson spheres as "a little joke" and commented that "you get to be famous only for the things you don't think are serious" [...]
To be fair, he later added this:
>in a later interview with students from The University of Edinburgh in 2018, he referred to the premise of the Dyson sphere as being "correct and uncontroversial".[13] In other interviews, while lamenting the naming of the object, Dyson commented that "the idea was a good one", and referred to his contribution to a paper on disassembling planets as a means of constructing one.
Thanks for pointing out those follow ups. Interesting stuff!
> correct and uncontraversial
From the original quote it is clear he was referring to the idea of aliens being detectable by infrared because they will absorb all of their sun's energy. Later in the same paragraph he says:
> Unfortunately I went on to speculate about possible ways of building a shell, for example by using the mass of Jupiter...
> These remarks about building a shell were only order-of-magnitude estimates, but were misunderstood by journalists and science-fiction writers as describing real objects. The essential idea of an advanced civilization emitting infrared radiation was already published by Olaf Stapledon in his science fiction novel Star Maker in 1937.
So the Dyson Sphere is a rhetorical vehicle to make an order-of-magnitude estimate, not a description of a thing that he thought could physically exist.
Full quote from the video cited before "the idea was a good one":
> science fiction writers got hold of this phrase and imagined it then to be a spherical rigid object. And the aliens would be living on some kind of artificial shell. a rigid structure surrounding a star. which wasn't exactly what I had in mind, but then in any case, that's become then a favorite object of science fiction writers. They call it the Dyson sphere, which was a name I don't altogether approve of, but anyway, I mean that's I'm stuck with it. But the idea was a good one.
Again he explicitly says this "wasn't exactly what I had in mind." This one hedges a bit more and could be interpreted as his saying the idea of a Dyson Sphere is a good one. He may have meant that in the sense of it being a good science fiction idea though, and he subsequently goes on to talk about that.
The Dyson Sphere is good for order-of-magnitude calculations about hypothetical aliens, and also for selling vapourware to the types of people who uncritically think that vapourware is real.
Have you read the paper itself, not just summaries of the idea? It's obvious from the way he wrote it, dripping in sarcasm. Talking about "Malthusian principles" and "Lebensraum", while hand waving away any common sense questions about how the mass of Jupiter would even be smeared into a sphere around the sun, just saying that he can conceive of it and therefore we should spend public money looking for it. He's having a lark.
Also, he literally said it was a joke, and was miffed that he was best know for something he didn't take seriously.
What do you think the limiting factor is? I don't see why we can't scale manufacturing of satellites up as far as we want. If we mine out a substantial fraction of the mass of the earth, we can go harvest asteroids or something.
You need to be able to harness enough raw material and energy to build something that can surround the sun. That does not exist in the solar system and we do not yet have the means to travel further out to collect, move, and construct such an incredibly huge structure. It seems like a fantasy.
The inner planets contain enough mass to create a shell of 1 AU radius with mass of 42 kg/m^2. That sounds like a plausible thickness and density for a sandwich of photovoltaics - GPUs - heat sinks.
You don't build a rigid shell of course, you build a swarm of free-floating satellites in a range of orbits.
The energy to build the system comes from the partial assembled system, plus some initial bootstrap energy. It grows exponentially. We seem to have enough today to build small factories in orbit.
The manufacturing scale comes from designing factory factories. They aren't that far in the future. Most factory machinery is made in factories which could be entirely automated, so you just need some robots to install machines into factories.
I was told ca. 2003 or so that because features on computer chips were getting smaller at some rate, and processor speed was getting faster at some other rate, that given exponential this or that I'd have tiny artificial haemo-goblins[1] bombing around my circulatory system that would make me swim like a fish under the sea for hours on end. But it turned out to be utter bullshit. Just like this.
Great. Now run the numbers to find the energy required to disassemble the planets and accelerating the pieces to their desired locations. For reference, it takes over 10 times of propellant and oxidant mass to put something in LEO.
The burned propellant and oxygen mass (as H2O and CO2) almost all ends up back in the atmosphere when you launch to LEO, so you can keep running electrolysis (powered by solar) to convert it back to fuel.
Sure, but if we're talking about solar engineering, that mass is going to be dispersed in orbit around the sun. You're not going to be reaccumulating that any time soon.
Also it's gravitationally unstable, like Dyson Rings, where as soon as you have any perturbance from the center means that the closer side is more attracted to the sun so it enters a feedback loop.
There are only so many people who can make satellites; there are only so many things to make satellites out of; and there are only so many orbits to put them in. There are only so many reasons why a person might want a satellite. There are only so many ways of placing satellites in orbit and each requires some amount of energy, and we have access to a finite amount of energy over time.
Finally, if we limited ourselves to earth-based raw materials, we would eventually reach a point where the remaining mass of the earth would have less gravitational effect on the satellite fleet than the fleet itself, which would have deleterious effects on the satellite fleet.
Seven reasons are intuitive; I’m sure there are many others.
People can build a factory that makes satellites. And then a factory that makes factories to make satellites.
There is plenty of material in the solar system (see my other response), and plenty of orbits, and launch capability can scale with energy harvested so the launch rate can grow exponentially.
Lots of people will probably decide they don't want any more satellites. But it only takes a few highly determined people to get it done anyway.
> Finally, if we limited ourselves to earth-based raw materials, we would eventually reach a point where the remaining mass of the earth would have less gravitational effect on the satellite fleet than the fleet itself, which would have deleterious effects on the satellite fleet.
The Earth's crust has an average thickness of about 15-20 km.
Practically we can only get at maybe the top 1-2 km, as drill bits start to fail the deeper you go.
The Earth's radius is 6,371 km.
So even if we could somehow dug up entire crust we can get to and flung it into orbit, that would barely be noticeable to anything in orbit.
After a few decades, you need to start replacing all the solar panels.
And the robot army being used to do the construction and resource extraction will likely have a much shorter lifespan. So needs to be self-replicating/repairing/recycling.
Yeah, but besides not having the physical amount of material available in the solar system, or the availability of any technology to transfer power generated to a destination where it can serve a meaningful purpose in the foreseeable future, or having the political climate or capital necessary for even initiating such an effort, or not being able to do so without severely kneecapping the habitability of our planet, there are aren't really any meaningful barriers that I can see.
Doesn't this risk some unforeseen effects on Earth or the rest of the solar system at that scale? Disruption of magnetic shield, some not yet known law of physics suddenly getting felt etc.?
Nameplate capacity needs a derate for availability, so you can drop it down to about 200GW(e) equivalent continuous power assuming we're making and deploying enough batteries to support it. More, obviously, if those panels are going to an equatorial desert, less if they're going to sunny Svalbard in the winter time.
A former NASA engineer with a PhD in space electronics who later worked at Google for 10 years wrote an article about why datacenters in space are very technically challenging:
I don't have any specialized knowledge of the physics but I saw an article suggesting the real reason for the push to build them in space is to hedge against political pushback preventing construction on Earth.
I can't find the original article but here is one about datacenter pushback:
But even if political pushback on Earth is the real reason, it still seems datacenters in space are extremely technically challenging/impossible to build.
The real reason is, Elon has SpaceX and xAI. He can create an illusion of synergy and orders of magnitude advancements to boost the market cap and pocket all the money. He realized long time ago you don't need to deliver to play the market cap game, in fact it's better if you are selling a story far in the future rather than a something you can deliver now.
We don’t even have a habitable structure in space when the ISS falls, there is no world in which space datacenters are a thing in the next 10, I’d argue even 30 years. People really need to ground themselves in reality.
Edit: okay Tiangong - but that is not a data center.
> We don’t even have a habitable structure in space
Silicon is way more forgiving than biology. This isn’t an argument for this proposal. But there is no technical connection between humans in space and data centers other than launch-cost synergies.
Okay, but a human being represents what, 200 W of power? The ISS has a crew of 3, so that's less than a beefy single user AI workstation at full tilt. If the question is whether it's practical to put 1-2 kW worth of computing power in orbit, the answer is obviously yes, but somehow I don't think that's what's meant by "datacenter in space".
I don't know, 10 years seems reasonable for development. There's not that much new technology that needs to be developed. Cooling and communications would just require minor changes to existing designs. Other systems may be able to be lifted wholesale with minimal integration. I think if there were obstacles to building data centers on the ground then we might see them in orbit within the next ten years.
The same things you are saying about data centers in space was said by similar people 10-15 years ago when Elon musk said SpaceX would have a man on Mars in 10-15 years.
We have had the tech to do it since the 90's, we just needed to invest into it.
Same thing with Elon Musks hyperloop, aka the atmospheric train (or vactrain) which has been an idea since 1799! And how far has Elon Musks boring company come to building even a test loop?
Yeah, in theory you could build a data center in space. But unless you have a background in the limitations of space engineering/design brings, you don't truly understand what you are saying. A single AI data center server rack takes up the same energy load of 0.3 to 1 international space station. So by saying Elon musk can reasonable achieve this, is wild to anyone who has done any engineering work with space based tech. Every solar panel generates heat, the racks generate heat, the data communication system generates, heat... Every kW of power generated and every kW of power consumes needs a radiator. And it's not like water cooling, you are trying to radiate heat off into a vacuum. That is a technical challenge and size, the amount of tons to orbit needed to do this... Let alone outside of low earth... Its a moonshot project for sure. And like I said above, Elon musk hasnt really followed through with any of his moonshots.
> A single AI data center server rack takes up the same energy load of 0.3 to 1 international space station.
The ISS is powered by eight Solar Array Wings. Each wing weighs about 1,050kg. The station also has two radiator wings with three radiator orbital replacement units weighing about 1,100kg each. That's about 15,000 kg total so if the ISS can power three racks, that's 5,000kg of payload per rack not including the rack or any other support structure, shielding, heat distribution like heat pipes, and so on.
Assuming a Falcon Heavy with 60,000 kg payload, that's 12 racks launched for about $100 million. That's basically tripling or quadrupling (at least) the cost of each rack, assuming that's the only extra cost and there's zero maintenance.
Falcon Heavy does not cost 100M when launching 60 metric tons.
At 60 metric tons, you're expending all cores and only getting to LEO. These probably shouldn't be in LEO because they don't need to be and you probably don't want to be expending cores for these launches if you care about cost.
The real problem typically isn't weight, it's volume. Can you fit all of that in that fairing? It's onli 13m long by 5m diameter...
His time estimates are notoriously, um, aggressive. But I think that's part of how his companies are able to accomplish so much. And they do, even if you're upset they haven't put a human on Mars fast enough or built one of his side quests.
"We specialize in making the impossible merely late"
> Cooling and communications would just require minor changes to existing designs.
"Minor" cooling changes, for a radically different operating environment that does not even have a temperature, is a perfect insulator for conduction and convection, and will actively heat things up via incoming radiation? "Minor" ? Citation very much lacking.
People always make this claim about world hunger elimination with no sources. Keep in mind we make more than enough calories to feed everyone on the planet many times over, it's a problem of distribution, of getting the food to the right areas and continuing cultivation for self sufficiency.
Even the most magnanimous allocators cannot defeat the realities of boots on the ground in terms of distribution. It is a very difficult problem that cannot be solved top down, the only solution we've seen is growth of economic activity via capitalistic means, lifting millions, billions out of poverty as Asia has done in the last century for example.
I argue that if you have literal hundreds of billions of hard cash to burn for stupid things like AI datacenters, you could afford to make the lives of millions of starving people not suck instead, pretty easily so. But to do that, you'd have to try, and that would mean actually doing something good for humanity. Can't have that as a billionaire.
If you're hellbent on arguing with a cult, it will be much cheaper to go down to your local Church of Scientology and try to convince them that their e-meter doesn't work.
As if company performance actually affected stock price when it comes to anything Elon Musk touches.
For fuck's sake, TSLA has a P/E of a whopping *392*. There is zero justification for how overvalued that stock is. In a sane world, I should be able to short it and 10x my money, but people are buying into Musk's hype on FSD, Robotaxi, and whatever the hell robot they're making. Even if you expected them to be successes, they'd need to 20x the company's entire revenue to justify the current market cap.
It's much easier to find a country or jurisdiction that doesn't care about a bunch of data centers vs launching them into space.
I don't get why we aren't building mixed use buildings, maybe the first floor can be retail and restaurants, the next two floors can be data centers, and then above that apartments.
Data centers don't do anything other than sit there and turn electricity into heat. They only emit nothing but heat (which could be useful to others in the building).
In America they have "temporary" jet turbines parked next to them burning gas inefficiently with limited oversight on pollution and noise because they are "temporary".
Mixed-use buildings with restaurants on the lower floors and residential on the upper floors are very common. Not sure what prisons have to do with anything.
> A former NASA engineer with a PhD in space electronics who later worked at Google for 10 years wrote an article about why datacenters in space are very technically challenging
It's curious that we live in a world in which I think the majority of people somehow think this ISN'T complicated.
Like, have we long since reached the point where technology is suitably advanced to average people that it seems like magic, where people can almost literally propose companies that just "conjure magic" and the average person thinks that's reasonable?
I can put things in a box that uses spooky electromagnetic waves to tickle water molecules to the point that they get hot and maybe boil off, given the chance? Sounds like magic to me
Nice article, the first one. I hope they try it, burn many billions of cash, and then fail. I also hope they don't spread radioactive material across the whole atmosphere when failing, though.
No, rockets landing themselves is just controlling the mechanism you use to have them take off, and builds on trust vectoring technology from 1970s jet fighters based on sound physics.
Figuring out how to radiate a lot of waste heat into a vacuum is fighting physics. Ordinarily we use a void on earth as a very effective _insulator_ to keep our hot drinks hot.
It's a matter of deploying it for cheaper or with fewer downsides than what can be done on earth. Launching things to space is expensive even with reusable rockets, and a single server blade would need a lot of accompanying tech to power it, cool it, and connect to other satellites and earth.
Right now only upsides an expensive satellite acting as a server node would be physical security and avoiding various local environmental laws and effects
Lower latency is a major one. And not having to buy land and water to power/cool it. Both are fairly limited as far as resources go, and gets exponentially expensive with competition.
The major downside is, of course, cost. In my opinion, this has never really stopped humans from building and scaling up things until the economies of scale work out.
> connect to other satellites and earth
If only there was a large number of satellites in low earth orbit and a company with expertise building these ;)
> And not having to buy land and water to power/cool it.
It's interesting that you bring that up as a benfit. If waterless cooling (i.e. closed cooling system) works in space, wouldn't it work even better on Earth?
You need to understand more of basic physics and thermodynamics. Fighting thermodynamics is a losing race by every measure of what we understand of the physical world.
This is a classic case of listing all the problems but none of the benefits. If you had horses and someone told you they had a Tesla, you'd be complaining that a Tesla requires you to dig minerals where a horse can just be born!
The difference is that it was mostly clueless people like Thunderf00t who said it was impossible, who nobody took seriously. I don’t remember that basically all relevant experts claimed it was near impossible with current technology. That’s the situation now.
There’s also fairly clear distinction with how insane Elons plan has become since the first plans he laid for Tesla and SpaceX and the plans he has now. He has clearly become a megalomaniac.
Funnily enough, some of the things people said about Tesla is coming true, because Elon simply got bored of making cars. It’s now plausible that Tesla may die as a car company which I would not have imagined a few years ago. They’re arguably not even winning the self driving and robotics race.
Currently SpaceX have managed to land the booster only, not the rocket itself, if you are thinking about Starship. And reusability of said rocket is also missing (collecting blown up pieces from the bottom of the ocean doesn't count!).
No, people made fun of Elon for years because he kept attempting it unsafely, skirting regulations and rules, and failing repeatedly in very public ways.
The idea itself was proven by NASA with the DC-X but the project was canceled due to funding. Now instead of having NASA run it we SpaceX pay more than we'd ever have paid NASA for the same thing.
He also said he could save the us a trillion dollars per year with DOGE, and basically just caused a lot data exfiltration and killed hundreds of thousands of people, without saving any money at all
> It(Solar) works, but it isn't somehow magically better than installing solar panels on the ground
Umm, if this is the point, I don't know whether to take rest of author's arguments seriously. Solar only works certain time of the day and certain period of year on land.
Also there is so limited calculations for the numbers in the article, while the article throws of numbers left and right.
Elon can't legally financially entangle Tesla to SpaceX due to Tesla being a public company, so his hands are tied.
Tesla is clearly benefiting from protectionism and its sales would collapse if BYD were allowed to openly sell in the US. Most people just want affordable, maintainable and reliable cars.
> Elon can't legally financially entangle Tesla to SpaceX due to Tesla being a public company, so his hands are tied.
He absolutely could do it, just like he did when Tesla bought SolarCity. It just isn’t as easy when one of the companies is public than when both are private.
We're witnessing a bailout and downloading of costs, at scale. Whether or not one buys into whatever the vision of these companies are - it's clear, there's interdealing.
Tesla theoretically now owns a chunk of xAI... whose valuation will no doubt increase due to the internalized SpaceX acquisition. Append to this a future IPO, as discussed in the artice, presumably an eventual premium of 20-50% (reasonable, 14% purely for the ibankers when this will happen)... yields to an interesting bailout situation.
To me, the real question is why. The $2B from Tesla can't possibly move the needle for any party involved in this transaction. If this were to be work 50x as opposed to a potential 50% upside (hell, make it 2x for argument's sake) it still doesn't compute. So what's the actual reason.
> Tesla is clearly benefiting from protectionism and its sales would collapse if BYD were allowed to openly sell in the US
So would most of EU car makers in Europe. China is not playing by the same rules and everyone with car manufacturing domestically is slamming them with tariffs.
How isn't China playing by the same rules? Every country subsidises and supports industry it thinks is important, surely nothing would stop Germany from investing into Volkswagen and BMW or the US from investing into Ford the same way China invests into BYD?
> Elon can't legally financially entangle Tesla to SpaceX
Bill Ackman has proposed taking SpaceX public by merging it with his Pershing Square SPARC Holdings, distributing 0.5 Special Purpose Acquisition Rights (SPARs) to Tesla shareholders for each share held. Each SPAR would be exercisable for two shares of SpaceX, aimed at enabling a 100% common stock capitalization without traditional underwriting fees or dilutive warrants.
With SpaceX IPO set to be one of the biggest of all time, this could have a pretty gnarly financial engineering impact on both companies -- especially if the short interest (direct or through derivatives) remains large.
Why would SpaceX go public? They already have a robust enough private market to give liquidity to all of their employees and shareholders who want it. They can get more private investment.
Going public would add a lot of hassle for little to no gain (and probably a negative of having to reveal their finances).
It has been widely reported for weeks that SpaceX is planning to go public in a few months. The reason is they have big plans to run a vast network of AI servers in orbit and will need to raise a massive amount of funding. xAI merger fits with that plan. I'd assume SpaceX still plans to go public.
Was ignored on HN but here's an article explaining:
it wasn't ignored on HN, there were many articles correctly noting that building data centers in space is a stupid stupid idea because cooling things there is infeasible
Was doing some back of the envelope math with chatGPT so take it with a grain of salt, but it sounds like in ideal conditions a radiator of 1m square could dissipate 300w. If this is the case, then it seems like you could approach a viable solution if putting stuff in space was free. What i can't figure out is how the cost of launch makes sense and what the benefit over building it on the ground could be
Google, Blue Origin and at least 5 other smaller companies have announced plans to build data centers in space. My understanding is the cooling issue is not the show stopper you assume.
yup, bezos said "we will be able to beat the cost of terrestrial data centers in space in the next couple of decades". presumably this means they'll need huge ass radiators, so its all about bringing down launch costs since they'll need to increase mass.
AI datacenters are bottlenecked by power, bandwidth, cooling, and maintenance. Ok sure maybe the Sun provides ample power, but if you are in LEO, you still have to deal with Earth's shadow, which means batteries, which means weight. Bandwidth you have via starlink, fine. But cooling in space is not trivial. And maintenance is out, unless they are also planning some kooky docking astromech satellite repair robot ecosystem.
Maybe the Olney's lesions are starting to take their toll.
The shadow thing can be solved by using a sun-synchronous orbit. See for example the TRACE solar observation satellite, which used a dawn/dusk orbit to maintain a constant view of the sun.
Every telco satellite can cool its electronics. However, more than a few kW is difficult. The ISS has around 100kW and is huge and in a shadow half the time.
The cooling is the bit where I'm lost on, but it will be interesting to see what they pull off. It feels like everyone forgets Elon hires very smart people to work on these problems, it's not all figured out by Elon Musk solely.
Google, Blue Origin and a bunch of other companies have announced plans for data centers in space. I don't think cooling is the showstopper some assume.
Good call out, and really interesting. SpaceX being the cheapest way to get things into space, it seems like SpaceX is about to become extremely lucrative.
He's broken pretty much all the other financial rules.... for example, the amount of blatant self-dealing he gets away with is staggering.
As long as the consequences of his actions continue to increase the paper value for investors, regulations don't really have teeth because there aren't damages. So the snowball gets bigger and the process repeats.
I've been thinking about this recently as I hear it often. Would people who want to buy a car in the Tesla price range really choose a slightly cheaper Chinese EV if those were available?
Personally I have a hard time believing this. But even if you had similarly priced Chinese options, I would guess the main reason for buying a Tesla is not just because you want an EV. While a Tesla will be a reliable baseline EV, surely the reason you (or at least I) would buy one is for the supervised self-driving feature.
BYD are just affordable and maybe reliable, regarding maintenance their spares are hard to come by and are almost as hard to work with as Tesla and other brands.
I've done plenty of work on my own Tesla. It's not hard to work on at all. Parts are not even very difficult. There are plenty of 3rd party shops (such as one I went to when I needed to replace my windshield.) I really wonder why people continue to think this. It's not 2016 any more.
Tesla body work is extremely expensive. Aluminum, extensive welding instead of fasteners, substantially reduced modularity due to castings, specialized tooling just off the top of my mind.
It's "ironic?" considering Tesla launching in China is what created the necessary supply chain to turn BYD into the powerhouse it is today. Tesla's greed will become their own demise.
Tesla cars made in Shanghai are sold in Europe and other places. That is helping them be competitive and they haven't had much price pressure until recently. Just because the Chinese have their own internal competition and deflation which drove their prices down aggressively doesn't mean it was a bad idea to build there. Also the idea the Chinese couldn't figure it out without an American company coming there first to show them is pretty silly.
Tesla Shanghai opened in 2019
BYD made their first hybrid in 2008 and they were a battery company since the 90s
Starship has a large number of critical milestones coming: Can it land and quickly reuse the upper stage? If not, it can't make refueling flights without building a dozen or two starships. Can it carry the full specified payload? If not, it can't even try to refuel in orbit. If it can't refuel in orbit, it can't go beyond earth orbit. Etc.
Everything has to go right or it will be irrelevant before it works.
> Everything has to go right or it will be irrelevant before it works.
Starship is not all of SpaceX. Saying, maybe because one hates Musk, that SpaceX is going to become irrelevant is wishful thinking.
In 2025 SpaceX launched more rockets into space than the entire world ever sent in a year up to 2022, something crazy like that.
Then out of, what, 14 000 active satellites in space more than half have been launched by SpaceX.
SpaceX is, so far, the biggest space success story of the history of the human race (and GP is right in saying that SpaceX is now a national security matter for the US).
Model S was the most successful EV. If you think cybercab is the vehicle of the future, look at the timeline of the only robo taxi in commerce in the US.
Everything has to go right with that, or cybercab will be irrelevant before it works. Same deal. Same bullshitter.
Model S was successful until Model 3/Y blew it out of the water. Waymo’s timeline is not relevant because they lose money on every car and every deployment. Tesla’s the only financially successful developer of self driving. They can scale it up much faster.In fact, instead of making $5k per car produced, cybercab will net them $50k per car per year.
Tesla remains competitive in China, which can't be said of European EVs. Chinese ADAS are much better than European ones but still far behind FSD.
To bring the discussion back on topic: $50k/year or ~$250k over the course of the vehicle's lifetime, instead of $5k for a singular sale event, is why the path for the company is crystal clear. Cybercab is the same kind of step for Tesla as the Model 3 was back in 2017.
I’ll grant you that it could be, and I’m betting it won’t while you are betting it will. The future is now obvious to fsd14 and robotaxi users. Failure is no longer likely.
This would literally only be said by a person who hasn’t tried fsd14 so do yourself a favor and go for a test ride of a Tesla and Audi/Chevrolet and report back. They are not comparable.
They sure as hell do not. SuperCruise only worked in pre mapped areas and bails whenever there's construction or deviation to plan. It's analogous to Tesla AP2 at best.
How vital is it really to national security? Starlink will have competition from Amazon Leo in the next few months. And while SpaceX is obviously in the lead in launch capability with Starship, there are multiple launch providers capable of providing roughly the same services the Falcon 9 and Heavy provide today.
The same services as Falcon 9 are 20x the cost and launch 1/20th as much as well. That's like producing hand made good in America versus via a manufacturing line in China.
Haven't you heard? Tesla is pivoting to building humanoid robots instead. They haven't sold a single one, but it toootally warrants retooling their car factories, pinky promise!
Oh c'mon now. Damn model 3 and model S I have driven were considerably lower quality interiors than an ass end Citroen or Fiat. The Model S, a 2023 model the doors didn't even fit properly. And that was all Europe.
As for FSD, nope. Unless you redefine the word reliable.
Edit: I owned a 2018 Model S as well. Literally the worst fucking car I have ever owned or driven.
> BMW actually has a reasonable control surface though, not a grand user interface experiment by some crack heads.
Really? It's one thing to hate Elon Musk, but you're talking about a lot of brilliant engineers who worked on these cars, everything from the components to the software. It's uneeded low blow just because you don't like Elon Musk.
The UX is a mess. Why does the car always label the trunk as open rather than have a button that I press to open it?
Why does cruise control sometimes change to the speed limit and sometimes not?
Why does auto lane change sometimes need me to start the manoeuvre and sometimes not? If I guess wrong and start the lane change myself, all autopilot just disengages suddenly.
I have to proove that I'm holding the wheel by wiggling it from time to time, but if I accidentally wiggle too hard it disengages. Why not have a sensor or use the cameras to detect if I'm holding the wheel?
My son didn't shut the back door properly. I started driving and the car started binging. It didn't tell me why it was binging until I put it in park and looked at the pretty 3d representation of the car, then noticed that the door was open.
Maybe if I drove more regularly I would get used to all this stuff. The car was borrowed and I gave it back.
I’m glad you found a place to get these complaints off your chest, but these are kind of hilarious. the button says “open trunk”. It’s a verb. If this is your complaint then lmao have you not seen what other OEM software looks like? Door open doesn’t just ding, it shows a warning with plain english explanation and an icon.
For the rest of your complaints you can mostly thank the overzealous EU/unece regulation which limits steering torque and requires intervention. FSD has none of those concerns, it just drives and does not require torque on the wheel.
That is what they want you to think it isn't too big to fail there are plenty of competitors with much stronger engineers and principles than this grifter.
Of course it isn't "too big to fail". Even banks aren't. Despite recent history large banks have failed often throughout history. There's no such thing. It may take down the supporting sovereign government (Dutch East Indies) but life goes on and new political orgs appear. People be people.
Too big to fail is a very recent modern myth. Go back 100+ years and lots of banks failed leading into the Great Depression.
Right. You do have a point, and I think Dutch East Indies is a good example, but I feel this is discussing semantics. Too big to fail, I interpret in this situation as the government having a strategic reason to keep it afloat so it will probably prop it up in case something goes wrong. This makes it have a much more stable position.
Let’s be honest - this is just a way to prop up Twitter/X. It makes SpaceX shareholders subsidize X, and also American taxpayers who are giving contracts to SpaceX for highly sensitive things. The government should ideally refuse to give SpaceX work unless it unwinds this.
Because Twitter/X is distorting our politics (with ann unbalanced scheme of censorship / amplification / suppression) and destroying the country by mainstreaming far right supremacist politics. Twitter/X does not deserve a single dollar of taxpayer money. If SpaceX is now part of that machine, it doesn’t deserve a single dollar either. I would rather pay more for alternatives and encourage their growth. I also look at any money given to this company as the equivalent of GOP campaign funding, so I feel it should be treated as illegal under the law.
Shouldn't the government be aiming to pay the lowest price for the best goods and services rather than using procurement as a way to promote or suppress certain political opinions?
The government is prevented from doing that by a little thing called the first amendment. "Mainstreaming far right supremacist politics" is just a hyperbolic way of saying he has politics you don't like and is exercising his freedom of the press by promoting it on the media platform he owns. Legally that is no different then the rights that every newspaper and TV station in the country has.
First of all, the current government doesn't give a shit about the first amendment and is successfully putting a chilling effect on it through various means. Both through illegally using government funding as a hammer to require independent companies to curtail their speech, or by using regulation.
Second, history will look back and realize that without taking into account the volume of your voice, you don't really have free speech in a way that matters. If you the person next to you can use a megaphone that is so loud that no one hears you, you effectively have no speech. A great many democracies implicitly realize this and thus have election spending limits tied to the number of supporters. The US, through it's lobby system, and through party affiliated control of third party networks, does not.
Musk is, indeed, allowed under the 1st to promote whatever he wants to promote. Him being a hypocrite about "free speech absolutism" is not a crime.
However, the current US administration appears to be actively violating the 1st and 5th in a bunch of ways, the 14th that one time, and making threats to wilfully violate the 2nd for people they don't like and the 22nd to get a third term. It is reasonable, not hyperbolic, to be concerned about Musk's support of this.
Actually the Trump administration is trying to strip legal status from people and deport them by way of an obscure law that gives the Secretary of State the discretion to do so if they deem those people a threat to the foreign policy goals of the US.
If these laws are still on the books when the next D administration takes over, they should use them against Elon, Thiel, etc - strip them of US citizenship, deport them, and nationalize their companies (followed with repealing those laws)
I disagree. He would be using taxpayer money to boost his preferred speech. And it is essentially campaign funding for the GOP. It should be treated as such.
If anything, I think this is actually the other way around - channeling crazy AI bubble money towards SpaceX, after the funding from goverment contracts has dried up. Twitter is just the icing on the cake.
Quite ingenious, you have to give Musk that. This is why he is making so much money.
It’s also a way to distract from the fact that alleged pedophile and rapist Elon had 3 underaged foreign nationals trafficked to him at the space x headquarters by convinced pedophile and rapist Jeffrey Epstein, per the Epstein files.
SpaceX is too big to fail for sure. If it goes bankrupt, it'll be broken down, and trimmed down to the succesful launching operation. But I don't think it's the reason it's buying xAI.
SpaceX buying xAi means that xAI shareholders are cashing in on its current high valuation. It makes it look like Musk is not very confident that xAI can navigate through the AI cycle, so he might as well sell it to rake in the profits.
But he still needs control over it because of the Tesla plan and in case something else happens in the AI field that he doesn't want to miss. So he's buying it with SpaceX, because he can, freeing some of SpaceX cash to pay himself and his xAI investors.
That he managed to bullshit SpaceX investors into buying xAI is pretty crazy. But I guess that's his main talent.
SpaceX is slated to go public some time this year - June IIRC
The biggest selling point /was/ that Musk was being managed there, he wasn't tinkering with SpaceX like Twitter or Tesla, and his foolhardy direction was kept out of the company.
BUT, like Tesla, Musk cannot help himself and is making SpaceX look like a very bad investment - tying his other interests with SpaceX, allegedly using SpaceX money as a "war chest" in his battles.
There is also a danger that investors will see xAI as politically dangerous, which will really hurt SpaceX IPO
They want to go public, but have to sell the hell out of it in the meantime.
I'll bet SpaceX financials aren't as great as some people think. Remember, Elon was the guy who tried to take Tesla private, and talked a lot of smack about how silly it is to be a public company. All of a sudden he wants SpaceX to go public?
Musk has a pattern here - he used Tesla the same way, diverting resources to xAI and treating it as a funding vehicle for other ventures. Once he started doing that, Tesla's financials got murky and harder to trust. Now he's doing it with SpaceX right before the IPO. For investors, that's not 'too big to fail' protection - it's a red flag that the company finances are entangled with his personal empire instead of focused on the core business.
> The biggest selling point /was/ that Musk was being managed there, he wasn't tinkering with SpaceX like Twitter or Tesla, and his foolhardy direction was kept out of the company
The biggest selling point to who? Definitely not wall street
What do you mean "let it fail?" SpaceX has the most profitable launch system in the world and now operates >50% of all satellites in orbit. They aren't exactly in need of a bailout.
When a company is operating at a scale where you are making orders of magnitude more orbital launches than NASA, operating a constellations of 10,000+ satellites, providing internet access to 10s of millions of people and 1 army, has raised $10s of billions in private markets at valuations in the $100s of billions, then the burden of proof is on you claiming the opposite.
The proof is that they are continuing to launch more mass into orbit than any other entity on the planet - while holding share liquidity events for their employees multiple times a year where they buy back shares. Proof is that they charge a lower cost to orbit than any of their competitors and has done so for years now.
Their revenue from Starlink is slated to be bigger than the entire NASA budget this year.
NASA is still here. Unfortunately, NATA fell victim to enshitification by government contracting. NASA even if it wants to simply cannot today design and launch a rocket. :(
I am no fan of Musk the man. SpaceX is a strong company and Falcon is a solid vehicle. There is not a lot of competition, and NASA trying to in-source design and supply and construction of a new, reusable LEO rocket would be a complete nightmare.
I root for a competitive rocket market, but SpaceX is at the moment critical.
> The private sector is far better when it comes to money.
I've heard this a lot, but I've worked for BigCos and it seems like all they do is spend money, often superfluously. I've seen BigCos spend large quantities money on support contracts every year that haven't been used in more than a decade, or sending people on business trips across the country so they can dial into a meeting, or buying loads of equipment that sits dormant in warehouses for years and then is eventually sold off for pennies on the dollar.
I'm not convinced that they're better than the government with money allocation, I think they're just better at telling people they are.
I'm not talking specifically about SpaceX, although historically the cost of their rockets have been much lower than NASA. I'm being much more general. The public sector doesn't have the same incentives that private companies have, whether it's rockets or any other technology. It's sad, but it's the truth.
I wouldn't be too sure. Depends on NASAs mission profiles and a lot of factors. Falcon heavy can bring 26.7t to GTO in expendable mode and only 8t in reusable mode. Reusable cost of Falcon is US$97 million vs US$150 million expendable.
How much does it cost to develop and maintain the reusability? Is it worth the trade-offs in lower tons to orbit due to more weight? Is it worth it adjusting the payload into smaller units, including developing things like refueling in LEO?
Idk, I'm not on the inside doing those calculations...
> > Boeing is too important to fail as well but it’s been terrible as a shareholder
Your opinion on Boeing being terrible as a shareholder vis-a-vis Tesla would be completely reversed if dividends and capital gains of the 2 companies were to be offered in the form of miles to be flown on Boeing planes and miles on Teslas Uber/Taxi/Autonomous taxis instead of dollars
The absolute overperformance on the stock market that Tesla has enjoyed vis-a-vis Boeing is not rooted in a concrete and tangible quality of life improvement for citizens. Not American citizens, nor global citizens for that matter.
It is my opinion that for all public companies in which it is possible to do so government should mandate payment in kind to all shareholders and board members to prevent the excessive promotional , cult and all around BS aspect of marketing to take over and allow people to profit just by riding off those, and Musk is the GOAT at that.
Im not comparing it to Tesla, im comparing it to any normal successful company (apple, google, nvidia, Exxon, whatever).
Boeing is an anemic company that doesn’t innovate and it should have been allowed to bankrupt and break off into businesses that worked and actually competed for customers.
Why are we still supporting this person? His cars are being outclassed internationally and he's directly meddling in this countries politics. He spectacularly failed (or wasn't it blatantly misled) the CA government with regard to the tunneling, and damaged the public sector while shutting down oversight and regulatory bodies against his companies.
Where is the benefit? These awesome tech demos? It just screams charlatan to me on an epic scale. I see no reason a government shouldn't step in to assume control if its "too big to fail".
Twitter (X) owed $1.3B in debt every year in interest since Musk's takeover. This was before re-financing in a higher interest rate environment. The company was losing $200MM+ per year on ~$5B in revenue before the takeover, and there are reports that revenues have decreased by round 50%.
Best case scenario if we accept those numbers is that X makes $3B per year and about half of that goes immediately out the door in debt payments before paying a cent for the entire business to function.
However, if SpaceX acquires X, that ~$1.5B in interest is a fraction of the $8B In profits SpaceX is allegedly generating annually. Further, they can restructure the debt if it's SpaceX's debt, and not owned by X. Investors will be more likely to accept SpaceX shares as collateral than X.
> The company was losing $200MM+ per year on ~$5B in revenue before the takeover, and there are reports that revenues have decreased by round 50%.
X made a profit last year because they cut costs lower than the drop in ad revenue (which is also slowly recovering). The big question is if they will still be profitable in 2026 year without the US election driving big traffic numbers and ads.
You're right, wrote that from memory. It was EBITDA that surpassed anything Twitter previously had before purchasing it.
> Despite a revenue drop from $5 billion in 2021 to roughly $2.7 billion in 2024, the EBITDA margin surged from 13.6% to 46.3% due to drastic cost-cutting measures and restructuring
Banks released pricing to sell their debt. When the debt gets to valued near market value, it means it is essentially guaranteed to get paid back. The company was making much less money but was more profitable, see the other posters comment on EBITDA.
He buys twitter at an inflated valuation. Runs it to the ground to a much lower valuation of $9B. [1] Then, his company Xai buys Twitter at a $33B, inflating the valuation up. Then SpaceX merges with Xai for no particular reason, but is expected to IPO at a $1T+ in the upcoming years. [3]
I’m not that smart, but if I were, I would be thinking this is an extended way to move the losses from the Twitter purchase on to the public markets.
It also makes it impossible for Twitter/X to die, as it deserves. It is by far the most toxic mainstream social network. It has an overwhelming amount of far right supremacist content. So bad that it literally resulted in Vivek Ramaswamy, a gubernatorial candidate in Ohio, to quit Twitter/X - nearly 100% of replies to his posts were from far right racists.
Obviously advertisers have not been fans. And it is a dying business. But rather than it dying, Elon has found a clever (and probably illegal) way to make it so that SpaceX, which has national security importance, is going to prop up Twitter/X. Now our taxpayer dollars are paying for this outrageous social network to exist.
I find HN and the tech circles to be one of the main community pillars holding up X. None of my social friends use it anymore, but links absolutely abound here, and it seems like the standard line is to pretend Elon, Grok, all the one button revenge and child porn etc don’t exist. I truly can’t fathom the amount of not thinking about it it would take to keep using the platform.
I have a blocker set up in my browser to prevent accidental clicks and sending any traffic to them when I'm not careful to check a given HN link to a posting. I've never had an account there (nor any of the popular social media networks) but I don't want to send even my few clicks their way.
No it isn't, the sensible people you followed 5 years ago left and stopped posting. The "Your followers" feed is now just the terminally addicted and the angry demagogues.
It was easy to support SpaceX, despite the racist/sexist/authoritarian views of its owner, because he kept that nonsense out of the conversation.
X is not the same. Elon is actively spewing his ultraconservative views on that site.
Now that these are the same company, there's no separation. SpaceX is part of Musk's political mission now. No matter how cool the tech, I cannot morally support this company, and I hope, for the sake of society, it fails.
This announcement, right after the reveal that Elon Musk reached out to Jeffrey Epstein and tried to book a trip to Little St. James so that he could party with "girls", really doesn't bode well.
It's a shame you can't vote these people out, because I loved places like Twitter, and businesses like SpaceX and Tesla, but Elon Musk is a fascist who uses his power and influence to attack some of the most important pillars of our society.
Elon is moderate at best. If a democrat supported cutting the budget, having an actual border to the country, and keeping men out of women's bathrooms you'd get Elon.
Check out his X feed. He regularly posts unhinged things about white culture, western values, etc that are supremacist and often, lifted from other supremacists. In the last year he became far more radicalized towards the far right. If it was just the things you said I might agree.
> It also makes it impossible for Twitter/X to die, as it deserves. It is by far the most toxic mainstream social network. It has an overwhelming amount of far right supremacist content.
Twitter also has more (not total, but more) free speech than any other social networking site. For example, you are allowed to discuss empirical research on race, crime and IQ. That would get you rate limited or banned quickly on other websites, including HN.
ISS radiators run on water and ammonia. Think about how much a kg costs to lift to space and you'll see the economics of space data centers fall apart real fast. Plus, if the radiator springs a leak the satellite is scrap.
The point of the Starship program is to drop the cost of a kg going to space significantly - this isn't meant to be launched with rockets that aren't fully reusable.
I don't pretend to understand the thermodynamics of all of this to do an actual calculation, but note that the ISS spends half its time in the shadow of the earth, which these satellites would not do.
You would put these in polar orbits so they are always facing the Sun. Basically the longitude would follow the Sun (or the terminator line, whichever you prefer), and the latitude would oscillate from 90°N to 90°S and back every 24 hours.
No. Otherwise how would you power them? We could use nuclear power methods, like we did in the Voyagers for instance. But the press release doesn’t mention that and, for a constellation of satellites around the earth, it would be a terrible idea.
Radiator size scales linearly with power but, crucially, coolant power, pumps, etc do not.
Imagine the capillary/friction losses, the force required, and the energy use(!) required to pump ammonia through a football-field sized radiator panel.
Moving electricity long distance is a lot easier than moving coolant long distances, which puts a soft limit on the reasonable size of the solar array of these satellites. But as long as you stay below that and pick a reasonable orbit it's indeed not too bad, you just have to properly plan for it
It does if you don't turn off the heat source every 30 minutes or so. Since the "datacenters" are targeted at sun synchronous orbits they have 24/7 heat issues. And they convert pretty much all collected energy into heat as well (and some data, but that's negligible). Those GPUs are not magically not generating heat.
Wouldn't the panels themselves need cooling too? The ones on earth generate heat while being in the sun.
There are commercial systems that can use open loop cooling (i.e. spray water) to improve efficiency of the panel by keeping the panel at a optimal temp of ~25C and the more expensive closed loop systems with active cooling recovers additional energy from the heat by circulating water like a solar heater in the panel back.
Twenty-five years after the ISS began operations in low Earth orbit, a new generation of advanced solar cells from Spectrolab, twice as efficient as their predecessors, are supplementing the existing arrays to allow the ISS to continue to operate to 2030 and beyond. Eight new arrays, known as iROSAs (ISS Roll-Out Solar Arrays) are being installed on the ISS in orbit.
The new arrays use multi-junction compound semiconductor solar cells from Spectrolab. These cells cost something like 500 times as much per watt as modern silicon solar cells, and they only produce about 50% more power per unit area. On top of that, the materials that Spectrolab cells are made of are inherently rare. Anyone talking about scaling solar to terawatts has to rely on silicon or maybe perovskite materials (but those are still experimental).
I'm not the best person to make that case as I can only speculate (land cost, permitting, latency, etc). /Shrug
In all the conversations I've seen play out on hacker news about compute in space, what comes up every time is "it's unviable because cooling is so inefficient".
Which got me thinking, what if cooling needs dropped by orders of magnitude? Then I learned about photonic chips and spintronics.
If you're considering only viability, the obvious concern would be cooling, yes; because increasingly large radiative cooling systems dominate launch costs because of all the liquid you need to boost into orbit. And one 100MW installation would be 500 times the largest solar power/radiative cooling system we've ever launched, which is the ISS. So get that down 2 orders of magnitude and you're within the realm of something we _know_ is possible to do instead of something we can _speculate_ is possible.
After that frankly society-destabilizing miracle of inventing competitive photonic processing, your goal of operating data centers in space becomes a tractable economic problem:
Pros:
- You get a continuous 1.37 kW/m^2 instead of an intermittent 1.0 kW/m^2
- Any reasonable spatial volume is essentially zero-cost
Cons:
- Small latency disadvantage
- You have to launch all of your hardware into polar orbit
- On-site servicing becomes another economic problem
So it's totally reasonable to expect the conversation to revolve around cooling, because we know SpaceX can probably direct around $1T into converting methane into delta-V to make the economics work, but the cooling issue is the difference between maybe getting one DC up for that kind of money, or 100 DCs.
Well, the primary limit on computation today is heat dissipation (the "power wall"). You either need to limit power so your phone or laptop doesn't destroy itself, or pay more to evacuate heat produced by the chips in your data center, which has its own efficiency curve.
If we suddenly lose 2 orders of magnitude of heat produced by our chips, that means we can fit 2 orders of magnitude more compute in the same volume. That is going to be destabilizing in some way, at the very least because you will get the same amount of compute in 1% the data center square footage of today; alternatively, you will get 100-900x the compute in today's data center footprint. That's like going from dial-up to fiber.
you think we don't have enough space on earth for a few buildings? this seems like a purely western cope. China seems perfectly able to build out large infrastructure projects with a land area smaller than that of the continentenal USA
You're presumably looking at a source that's including water area. When talking about land area, China > USA > Canada. (As opposed to when including water area, Canada > USA > China)
5,000 Starship launches to match the solar/heat budget of the 10GW "Stargate" OpenAI datacenter. The Falcon 9 family has achieved over 600 launches.
The ISS power/heat budget is like 240,000 BTU/hr. That’s equivalent to half of an Nvidia GB200 NVL72 rack. So two international space stations per rack. Or about 160,000 international space stations to cool the 10GW “Stargate” datacenter that OpenAI’s building in Abilene. There are 10,000 starlink satellites.
Starship could probably carry 250-300 of the new V2 Mini satellites which are supposed to have a power/heat budget of like 8kW. That's how I got 5,000 Starship launches to match OpenAI’s datacenter.
Weight seems less of an issue than size. 83,000 NVL72’s would weigh 270 million lbs or 20% of the lift capacity of 5000 starship launches. Leaving 80% for the rest of the satellite mass, which seems perhaps reasonable.
Elon's napkin math is definitely off though, by over an order of magnitude. "a million tons per year of satellites generating 100 kW of compute power per ton" The NVL72's use 74kW per ton. But that's just the compute, without including the rest of the fucking satellite (solar panels and radiators). So that estimate is complete garbage.
One note: If you could afford to send up one of your own personal satellites, it would be extremely difficult for the FBI to raid.
you put the radiators and the rest of the satellite within the shade of the solar panels, you can still make the area arbitrarily large
EDIT: people continue downvoting and replying with irrelevant retorts, so I'll add in some calculations
Let's assume
1. cheap 18% efficient solar panels (though much better can be achieved with multijunction and quantum-cutting phosphors)
2. simplistic 1360 W/m^2 sunlight orthogonal to the sun
3. an abstract input Area Ain of solar panels (pretend its a square area: Ain = L ^ 2)
4. The amount of heat generated on the solar panels (100%-18%) * Ain * 1360 W / m ^ 2, the electrical energy being 18% * Ain * 1360 W / m ^ 2. The electrical energy will ultimately be converted to computational results and heat by the satellite compute. So the radiative cooling (only option in space) must dissipate 100% of the incoming solar energy: the 1360 W / m^2 * Ain.
5. Lets make a pyramid with the square solar panel as a base, with the apex pointing away from the sun, we make sure the surface has high emissivity (roughly 1) in thermal infrared. Observe that such a pyramid has all sides in the shade of the sun. But it is low earth orbit so lets assume warm earth is occupying one hemisphere and we have to put thermal IR reflectors on the 2 pyramid sides facing earth, so the other 2 pyramid sides face actual cold space.
6. The area for a square based symmetric pyramid: we have
6.a. The area of the base Ain = L * L.
6.b. The area of the 4 sides 2 * L * sqrt( L ^ 2 / 4 + h ^ 2 )
6.c. The area of just 2 sides having output Area Aout = L * sqrt( L ^ 2 / 4 + h ^ 2 )
7. The 2 radiative sides not seeing the sun and not seeing the earth together have the area in 6.c and must dissipate L ^ 2 * 1360 W / m ^ 2 .
8. Hello Stefan-Boltzmann Law: for emissivity 1 we have the radiant exitance M = sigma * T ^ 4 (units W / m ^ 2 )
9. The total power exited through the 2 thermal radiating sides of the pyramid is then Aout * M
10. Select a desired temperature and solve for h / L (to stay dimensionless and get the ratio of the pyramid height to its base side length), lets run the satellite at 300 K = ~26 deg C just as an example.
11. If you solve this for h / L we get:
h / L = sqrt( ( 1360 W / m ^ 2 / (sigma * T ^ 4 ) ) ^ 2 - 1/4 )
12. Numerically for 300K target temperature we get: h/L = sqrt((1360 / (5.67 * 10^-8 * 300 ^ 4)) ^ 2 - 1/4)
= 2.91870351609271066729
13. So the pyramid height of "horribly poor cooling capability in space" would be a shocking 3 times the side length of the square solar panel array.
As a child I was obsessed with computer technology, and this will resonate with many of you: computer science is the poor man's science, as soon as a computer becomes available in the household, some children autodidactically educate themselves in programming etc. This is HN, a lot of programmers who followed the poor man's science path out of necessity. I had the opportunity to choose something else, I chose physics. No amount of programming and acquiring titles of software "engineer" will be a good substitute for physicists and engineers that actually had courses on the physical sciences, and the mathematics to follow the important historical deductions... It's very hard to explain this to the people who followed the path I had almost taken. And they downvote me because they didn't have the opportunity, courage or stamina to take the path I took, and so they blindly copy paste each others doomscrolled arguments.
Look I'm not an elon fanboy... but when I read people arguing that cooling considerations excludes this future, while I know you can set the temperature arbitrarily low but not below background temperature of the universe 4 K, then I simply explain that obviously the area can be made arbitrarily large, so the temperature can be chosen by the system designer. But hey the HN crowd prefers the layers of libraries and abstractions and made themselves an emulation of an emulation of an emulation of a pre-agreed reality as documented in datasheets and manuals, and is ultimately so removed from reality based communities like physics and physics engineering, that the "democracy" programmers opinions dominate...
So go ahead and give me some more downvotes ;)
If you like mnemonics for important constants: here's one for the Stefan Boltzman constant: 5.67 * 10^-8 W / m^2 / K ^ 4
thats 4 consecutive digits 5,6,7,8 ; comma or point after the first significant digit and the exponent 8 has a minus sign.
It all basically boils down to: in order to dissipate heat, you need something to dissipate heat into, e.g. air, liquid, etc. Even if you liquid cool the GPUs, where is the heat going to go?
On Earth, you can vent the heat into the atmosphere no problem, but in space, there's no atmosphere to vent to, so dissipating heat becomes a very, very difficult problem to solve. You can use radiators to an extent, but again, because no atmosphere, they're orders of magnitude less effective in space. So any kind of cooling array would have to be huge, and you'd also have to find some way to shade them, because you still have to deal with heat and other kinds of radiation coming from the Sun.
What you're describing is one of two mechanisms of shedding heat which is convection, heating up the environment. What the long comment above is describing is a _completely_ different mechanism, radiation, which is __more__ efficient in a vacuum. They are different things that you are mixing up.
that page has not a single calculation of radiative heat dissipation, seems like he pessimistically designed the satellite avoiding use of radiative cooling which forces him to employ a low operational duty cycle. Kind of a shame to be honest, given the high costs of launching satellites, his sat could have been on for a larger fraction of time...
It seems straightforward to you because you're ignoring everything that makes this not work.
Here's a big one: you can't put radiators in shadow because the coolant would freeze. ISS has system dedicated to making sure the radiators get just enough sunlight at any given time.
That helps with the heat from the sun problem, but not the radiation of heat from the GPUs. Those radiators would need to be unshaded by the solar panels, and would need to be enormous. Cooling stuff in atmosphere is far easier than in vacuum.
Not so. Look at the construction of JWST. One side is "hot", the other side is very, very cold.
I am highly skeptical about data centers in space, but radiators don't need to be unshaded. In fact, they benefit from the shade. This is also being done on the ISS.
"I meant they would need a clear path to open space not blocked by solar panels, but yes, a hot and cold side makes sense."
This is precisely why my didactic example above uses a convex shape, a pyramid. This guarantees each surface absorbs or radiates energy without having to take into account self-obscuring by satellite shape.
The goal of JWST is not to consume as much power as possible, and perform useful computations with it. A system not optimized for metric B but for metric A scores bad for metric B... great observation.
this makes no sense, the radiation of heat from the GPU's came from electrical energy, the electrical energy came from the efficient fraction of solar panel energy, the inefficient fraction being heating of the solar panel, the total amount of heat that needs to be dissipated is simply the total amount of energy incident on the solar panels.
at ~20% solar panel efficiency, we need 1.15 MW of optical power incident on the solar panels.
The required solar panel area becomes 1.15 * 10^6 W / 1.360 * 10^3 W / m ^ 2 = 846 m ^ 2.
thats about 30 m x 30 m.
From the center of the square solar panel array to the tip of the pyramid it would be 3x30m = 90 m.
An unprecedented feat? yes. But no physics is being violated here. The parts could be launched serially and then assembled in space. Thats a device that can pretrain from scratch LLaMa 3.1 in 16.8 days. It would have way to much memory for LLaMa 3.1: 16 x 8 x 192 GB = ~ 25 TB of GPU RAM. So this thing could pretrain much larger models, but would also train them slower than a LLaMa 3.1.
Once up there it enjoys free energy for as long as it survives, no competing on the electrical grid with normal industry, or domestic energy users, no slow cooking of the rivers and air around you, ...
Space is not empty. Satellites have to be boosted all the time because of drag. Massive panels would only worsen that. Once you boosters are empty the satellite is toast.
the point wasn't that a 1 m^2 solar panel could theoretically be kept reasonably cool at the cost of a miles long radiator... nono, the point was that you could attain any desirable temperature this way, arbitrarily close to 4K.
for a reasonable temperature (check my comment for updated calculations) the height of a square based pyramidal satellite would be about 3 times the side length of its base, quite reasonable indeed. Thats with the square base of the pyramid as solar panel facing the sun, and the top of the pyramid facing away, so all sides are in the shade of the base. I even halved my theoretical cooling power to keep calculations simple: to avoid a long confusing calculation of the heat emitted by earth, I handicapped my design so 2 of the pyramidal side surfaces are reflective (facing earth) and the remaining 2 side triangles of the pyramid are the only used thermal radiative cooling surfaces. Less pessimistic approaches are possible, but would make the calculation less didactic for the HN crowd.
It seems straightforward to you because you're ignoring everything that makes this not work.
Here's a big one: you can't put radiators in shadow because the coolant would freeze. ISS has system dedicated to making sure the radiators get just enough sunlight at any given time.
The ISS goes into Earth's shadow for ~45 minutes and then in the sun for 45 minutes, in 24/7 repeat;
this system would not be given such an orbit. Its trivial to decrease the cooling capacity of the radiators: just have an emissivity ~0 shade (say an aluminum foil) curtain obscure part of the radiator so that it locally sees itself instead of cold empty space. This would only happen during 2 short periods in the year.
The design issues of the ISS are totally different from this system.
"Satellites have to be boosted all the time because of drag."
On Low Earth Orbits (LEOs), sure, but the traces of atmosphere that cause the drag disappear quite fast with increasing altitude. At 1000 km, you will stay up for decades.
arbitrarily large means like measured in square km. Starcloud is talking about 4km x 4km area of solar panels and radiative cooling. (https://blogs.nvidia.com/blog/starcloud/)
Building this is definitely not trivial and not easy to make arbitrarily large.
When a physicist says arbitrarily large it could even be in a dimensionless sense. It doesn't matter how small or large the solar panel is:
for a 4 m x 4 m solar panel, the height of the pyramid would have to be 12 m to attain ~ 300 K on the radiator panels. Thats also the cold side for your compute.
for a 4 km x 4 km solar panel the height of the pyramid would be 12 km.
> you put the radiators and the rest of the satellite within the shade of the solar panels, you can still make the area arbitrarily large
The larger you make the area, the more solar energy you are collecting. More shade = more heat to radiate. You are not actually making the problem easier.
no the radiator planes are in the shade, so you can increase the height of a pyramidal shaped satellite for a constant solar panel base, and thus enjoy arbitrarily low rest temperatures, check my calculation which I added.
for a target temperature of 300K that would mean the pyramid height would be a bit less than 3 times higher than the square base side length h=3L.
I even handicapped my example by only counting heat radiation from 2 of the 4 panels, assuming the 2 others are simply reflective (to make the calculation of a nearby warm Earth irrelevant).
I'm not big on this subject, but I understand that heat transfer is difficult in space, because there's little to transfer to. If the solution is just making large radiators, then that means you're sending some big payloads full of radiators. Not to mention all the solar panels needed. I wanna live in sci-fi land too, but I don't see how it makes any sense compared to a terrestrial data center.
If someone has a design out there where this works and you can launch it economically on a rocket today, I wanna see that. And then I wanna compare it to the cost of setting up some data centers on earth (which BTW, you can service in real time, it sounds like these will be one-and-done launches).
You can line the solar panels and radiators facing away from each other, and the radiators would take up less surface area. I think maybe the tricky part would be the weight of water + pipes to move heat from the compute to the radiators.
There's plenty of space in space, but there isn't plenty of space in rocket fairings, nor is there plenty of lift capacity for an unlimited amount of radiators.
It's such bullshit that we've decided this moron and others in his cohort can unilaterally reallocate such vast portions of humanity's labor at their whims.
This is an extremely stupid idea, but because of our shared delusion of capitalism and the idea that wealth accumulation at the top should be effectively limitless, this guy gets to screw around and divert actual human labor towards insane and useless projects like this rather than solving real world problems.
radiators can be made as long as desirable within the shade of the solar panels, hence the designer can pracitically set arbitrarily low temperatures above the background temperature of the universe.
Radiators can shadow each other, so that puts some kind of limit on the size of the individual satellite (which limits the size of training run it can be used for, but I guess the goal for these is mostly inference anyway). More seriously, heat conduction is an issue: If the radiator is too long, heat won't get from its base to its tip fast enough. Using fluid is possible, but adds another system that can fail. If nothing else, increasing the size of the radiator means more mass that needs to be launched into space.
"Radiators can shadow each other," this is precisely why I chose a convex shape, that was not an accident, I chose a pyramid just because its obvious that the 4 triangular sides can be kept in the shade with respect to the sun, and their area can be made arbitrarily large by increasing the height of the pyramid for a constant base. A convex shape guarantees that no part of the surface can appear in the hemispherical view of any other part of the surface.
The only size limit is technological / economical.
In practice h = 3xL where L was the square base side length, suffices to keep the temperature below 300K.
If heat conduction can't be managed with thermosiphons / heat pipes / cooling loops on the satellite, why would it be possible on earth? Think of a small scale satellite with pyramidal sats roughly h = 3L, but L could be much smaller, do you actually see any issue with heat conduction? scaling up just means placing more of the small pyramidal sats.
Shading does work; JWST does this. However I don't see how you can make it work for satellite data centers. You would constantly be engaging attitude control as you realigned the panels to keep the radiators in shade. You'd run out of thruster fuel so fast you'd get like a month out of each satellite
Radiators can only be made as long as desirable because there's gravity for the fluid inside to go back down once it condenses. Even seen those copper heat pipes in your PC radiator?
these same comments pop up every time someone brings up satellite data-centers where people just assume the only way of dissipating heat is through convection with the environment.
No, we just "assume" (i.e. know) that radiation in a vacuum is a really bad way of dissipating heat, to the point that we use vacuum as a very effective insulator on earth.
Yes, you can overcome this with enough radiator area. Which costs money, and adds weight and space, which costs more money.
Nobody is saying the idea of data centers in space is impossible. It's obviously very possible. But it doesn't make even the slightest bit of economic sense. Everything gets way, way harder and there's no upside.
> No, we just "assume" (i.e. know) that radiation in a vacuum is a really bad way of dissipating heat, to the point that we use vacuum as a very effective insulator on earth.
In space or vacuum radiation is the best way to dissipate heat, since it's the only way.
I believe the reason the common person assumes thermal radiation is a very poor way of shedding heat is because of 2 factoids commonly known:
1. People think they know how a vacuum flask / dewar works.
2. People understand that in earthly conditions (inside a building, or under our atmosphere) thermal radiation is insignificant compared to conduction and convection.
But they don't take into account that:
1) Vacuum flasks / dewars use a vacuum for thermal insulation. Yes and they mirror the glass (emissivity nearer to ~0) precisely because thermal radiation would occur otherwise. They try their best to eliminate thermal radiation, a system optimized to eliminate thermal radiation is not a great example of how to effectively use thermal radiation to conduct heat. The thermal radiation panels would be optimized for emissivity 1, the opposite of whats inside the vacuum flask.
2) In a building or under an atmosphere a room temperature object is in fact shedding heat very quickly by thermal radiation, but so are the walls and other room temperature objects around you, they are reheating you with their thermal radiation. The net effect is small, in these earthly conditions, but in a satellite the temperature of the environment faced by the radiating surfaces is 4K, not a temperature similar to the object you are trying to keep cool.
People take the small net effect of thermal radiation in rooms etc, and the slow heat conduction through a vacuum flasks walls as representative for thermal radiation panels facing cold empty space, which is the mistake.
Well no, it’s because conduction/convection into a fluid is so much more effective.
Just look at a car. Maybe half a square meter of “radiator” is enough to dissipate hundreds of kW of heat, because it can dump it into a convenient mass of fluid. That’s way more heat than the ISS’s radiators handle, and three orders of magnitude less area.
Or do a simple experiment at home. Light a match. Hold your finger near it. Then put your finger in the flame. How much faster did the heat transfer when you made contact? Enough to go from feeling mildly warm to causing injury.
but thats what you don't get: conduction / convection on the ground is ultimately still radiation to space: you heat up our rivers, soils, atmosphere and the heat is eventually shed... by thermal radiation.
its not exactly good advertisement for conductive or convective heat transfer if its really employing thermal radiation under the hood!
but do you want big tech to shit where you eat? or do you want them to go to the bathroom upstairs?
At some point I'm thinking the large resistance to the idea I am seeing in a forum populated with programmers is the salivation-inducing idea that all that datacenter hardware will eventually get sold for less and less, but if we launch them to space there won't be any cheap devalued datacenter hardware to put in their man-caves.
Additional radiator area means bigger spacecraft, implies more challenge with attitude control. Lower down you get more drag so you use propellant to keep yourself up, higher up you have more debris and the large area means you need to frequently manoeuvre to avoid collisions. Making things bigger in space is not trivial! You can't just deploy arbitrarily large panels and expect everything to be fine.
heavier boats are also slower to accelerate or decelerate compared to smaller boats, does this mean we should ban container ships? having special orbits for megastructure lanes would seem a reasonable approach.
The radiators would be lighter compared to the solar panels, and slightly smaller surface area so you can line them back to back
I don't think dissipating heat would be an issue at all. The cost of launch I think is the main bottleneck, but cooling would just be a small overhead on the cost of energy. Not a fundamental problem.
If you solved this problem apply at nasa because they still haven't figured it out.
Either that or your talking out of your ass.
FYI a single modern rack consumes twice the energy of the entire ISS, in a much much much much smaller package and you'll need thousands of them. You'd need 500-1000 sqm of radiator per rack and that alone would weight several tonnes...
You'll also have to actively cool down your gigantic solar panel array
eldenring is slightly wrong: for reasonable temperatures the area of the radiating panels would have to be a bit more than 3 times the area of the solar panel, otherwise theres nothing wrong.
No need to apply at NASA, to the contrary, if you don't believe in Stefan Boltzmann law, feel free to apply for a Nobel prize with your favorite crank theory in physics.
Whats your definition for reasonable temp? my envelope math tells me at 82 celsius (right before h100s start to throttle) you'd need about 1.5x the surface area for radiators. Not exactly back to back, but even 3x surface area is reasonable.
Also this assumes a flat surface on both sides. Another commenter in this thread brought up a pyramid shape which could work.
Finally, these gpus are design for earth data centers where power is limited and heat sinks are abundant. In the case of space data centers you can imagine we get better radiators or silicon that runs hotter. Crypto miners often run asics very hot.
I just don't understand why every time this topic is brought up, everyone on HN wants to die on the hill that cooling is not possible. It is?? the primary issue if you do the math is clearly the cost of launch.
I am the person who gave the pyramid shape as a didactic example (convexity means we can ignore self obscuration, and giving up 2 of the 4 triangular side surfaces of the pyramid allows me to ignore the presence of lukewarm earth).
My example is optimized not for minimal radiator surface area, but for minimal mathematical and physical knowledge required to understand feasibility.
Your numbers are different because you chose 82 C (355 K) instead of my 26 C (300 K).
Near normal operating temperatures hardware lifetime roughly doubles for every 10 deg C/K decrease in temperature (this does not hold indefinitely of course).
You still need to move the heat from the GPU to the radiator so my example of 26 deg C at the radiator just leaves a lot of room against criticism ;)
The distinction is that you don't need to compete for land area, that you don't cause local environmental damage by heating say a river or a lake, that you don't compete with meatbags for energy and heat dissipation rights.
Without eventually moving compute to space we are going to have compute infringe on the space, energy, heat dissipation rights of meatbags. Why welcome that?!?
> How efficient is thermal radiation through a vacuum again?
I provided the calculation for the pyramidal shape: if the base of a pyramid were a square solar panel with side length L, then for a target temperature of 300K (a typical back of envelope substitute for "room temperature") the height of the pyramid would have to be about 3 times the side length of the square base. Quite reasonable.
> Sure, it occurs, but what does the Stefan–Boltzmann law tell us about GPU clusters in space?
The Stefan-Boltzmann law tells us that whatever prevents us from putting GPU clusters in space, it's not the difficulty in shedding heat by thermal radiation that is supposedly stopping us.
Just picture a square based pyramid, like a pyramid from egypt, thats the rough shape. Lets pretend the bottom is square. For thermodynamic analysis, we can just pretend the scale is irrelevant, it could be 4 cm x 4 cm base or 4 km x 4 km base. Now stretch the pyramid so the height of the tip is 3 times the length of the sides of the square base, so 12 cm or 12 km in the random examples above.
If the base were a solar panel aimed perpendicular to sun, then the tip is facing away and all side triangles faces of the pyramid are in the shade.
I voluntarily give up heat dissipation area on 2 of the 4 triangular sides (just to make calculations easier, if we make them thermally reflective -emissivity 0-, we can't shed heat, but also don't absorb heat coming from lukewarm Earth).
The remaining 2 triangular sides will be large enough that the temperature of the triangular panels is kept below 300 K.
The panels also serve as the cold heat baths, i.e. the thermal sinks for the compute on board.
Not sure what you mean with wings, I intentionally chose a convex shape like a pyramid so that no part of the surface of the pyramid can see another part of the surface, so no self-obstruction for shedding heat etc...
If this doesn't answer your question, feel free to ask a new question so I understand what your actual question is.
The electrical power available for compute will be approximately 20% (efficiency of solar panels) times the area of the square base L ^ 2 times 1360 W / m ^ 2 .
The electrical power thus scales quadratically with the chosen side length, and thus linearly with the area of the square base.
This is why I come to this site. Obviously, Twitter's financials are struggling and theres more than a few people rich people who don't want to take the hit... but we can all drop that for a second to discuss the plausibility of data centers in space. Some links and comments I enjoyed:
* https://taranis.ie/datacenters-in-space-are-a-terrible-horrible-no-good-idea/
* https://en.wikipedia.org/wiki/Tiangong_space_station
* "Technically challenging", a nice way to say "impossible"
* "I’m not that smart, but if I were, I would be thinking this is an extended way to move the losses from the Twitter purchase on to the public markets."
* "ISS radiators run on water and ammonia. Think about how much a kg costs to lift to space and you'll see the economics of space data centers fall apart real fast. Plus, if the radiator springs a leak the satellite is scrap."
* "5,000 Starship launches to match the solar/heat budget of the 10GW "Stargate" OpenAI datacenter. The Falcon 9 family has achieved over 600 launches." [nerdsniper]
* "No, we just "assume" (i.e. know) that radiation in a vacuum is a really bad way of dissipating heat, to the point that we use vacuum as a very effective insulator on earth."
* "World's Best At Surfing A Temporary Hyperinflation Wave is not a life goal to really be proud of tbh"
I guess props to scamming Compaq into making a large investment that didn't pan out. He did personally make money so I guess win for him.
>In an effort to woo investors, Elon Musk built a large casing around a standard computer to give the impression that Zip2 was powered by a supercomputer.
>PayPal
Huh? He didn't found Paypal, his company was acquired by Paypal. You might as well give him credit for eBay while you're at it. Paypal released their first digital wallet in 1999. They acquired x.com (and Musk) in 2000. Paypal itself was then acquired by eBay in 2002.
>Tesla
Investor, not founder.
>SpaceX
Yup, props here.
>Grok/xAI
Hasn't made a penny, no signs it had any path to profitability, which is why it was shoved into Space-X to cover his personal losses.
Taking Tesla from where it was (an overpriced prototype) to what it is now did take some skill. He wasn't some passive investor who put money in and didn't do anything. The rest for sure he was gotten credit that isn't earned.
hey now. Lets not forget when Elon had Grok creating CSAM and sexually explicit material of nonconsenting women. Truely an... achievement? Surely it will propel humanity forward.
He can take out a full page Wall Street Journal ad tomorrow that says “I created hyperloop to kill CA HSR” and it will have no effect on the fact that CA HSR’s failure is 100% the fault of CA’s own dysfunction.
Yeah that’s where I’m confused about this “conspiracy theory” stuff. It’s common knowledge that Musk wanted hyperloop to undermine the high speed rail project and also it later failed. Aside from a single HN comment I have never seen anyone attribute him with that much influence on the thing, so it is bizarre to see someone talking like there’s some sort of common conspiracy theory that Elon Musk controls trains or whatever. As far as I know pretty much nobody believes that.
There is no conspiracy theory, that aside the link does not indicate that there is one? “Vaguely accurate” does not mean “untrue”, and Vance is clear that he is talking about his personal interpretation of what Elon Musk is documented to have said, which he does not refute.
I like the idea that “he didn’t say that” and “he did say that but a different guy feels like he probably meant something else” are so obviously equivalent that skepticism of that notion constitutes a ‘conspiracy theory’.
That aside I like that the guy whose opinion should be treated as indisputable fact said that he thinks that there hasn’t been any high speed rail built globally in the past decade, which is not even remotely true. Obviously if he meant to say in the US he would have said so, since his next sentence was praise of Musk’s world-wide achievements.
I suppose it’s possible that Vance either doesn’t know anything about high speed rail or was in such a rush to extoll the virtues of the CEO of Tesla that he just sort of blurted something out to make Musk look good?
The full quote is “vaguely accurate but a disingenuous take”. And “Disingenuous” means “misleading/dishonest/untruthful/insincere/unfair”.
> Obviously if he meant to say in the US he would have said so
Come on, from the context it is clear that Vance means the US and specifically California. He also says “we” in the sentence “In all this time we've been talking about high-speed rail” and does not mean Chinese/Japanese/French having this discussion.
Disingenuous speaks to the motivation of the speaker, not the veracity of information on its own. Vance says that in his opinion that that particular interpretation of the factual information is disingenuous. As you pointed out, it can mean “unfair” which is not the same thing as untrue. Dude had an opportunity to say “that’s not true” and didn’t do that.
You’ve sort of just added “I feel like Vance meant something other than what he said” on top of Vance saying he felt like Musk meant something other than what he said. There isn’t a number of layers of “I feel…/he feels…” that you can pile onto a statement that will equal “he did not say the thing that he is quoted as having said”
Your contention is that by “accurate” he meant “inaccurate” and that he sees Elon Musk as being a global phenomenon and high speed rail as a… thing that’s local to the US? That is notable for its… absence?
Seems like “yeah that’s what he said but in my opinion you’re being mean to my friend” is more likely than a professional writer not knowing how to say “that’s not true”
It is patently clear what Musk meant, the guy isn’t famous for nuance. That aside I don’t find it difficult to picture the man that publicly claims that he personally elected the president thinking that he could sabotage a rail project. Now, I can’t know for sure that he believes that his Hyperloop pitch was responsible for the failure of the CA high speed rail project but if I had to make a bet about that…
hmm Tesla shipped millions of cars SpaceX launches 90% of space payloads, Starlink is working well. Thats hard to categorize as never delivered on any of his projects
The crucial thing is that Tesla's valuation has the hype projects baked in. The fact that it never delivered self driving or a robotaxi fleet and is now being saved solely by an import ban on Chinese EVs means that any success he had with Tesla is now an illusion.
There is another way to view this. FSD plays fast and loose because they are constantly iterating. The culture at Musk co is that if you dont' keep pushing updates you are in trouble so do we really want to trust that each of his numerous updates are truly tested? This guy is a pathological liar after all. How many lawsuits are they dealing with now?
Supercruise only runs on pre mapped routes. If my life is on the line, I'd rather take the pre mapped routes and supercruise design is better at preventing people playing games to defeat the system (ex.shoving an orange in the steering wheel) so I know that others using the system on the road are following the system guidelines.
Supercruise may not do everything FSD does but it cuts out a large portion of the "fatigue" portion of driving and as a result can be highly trusted value add.
The projects promised to be life altering for all mankind, they ended up being not even life altering for super rich Americans considering that Teslas are just EVs which without FSD are just regular cars with a different propellent that were made for political purposes and virtue signaling
The EV revolution has always been something almost dystopic : Trillions of dollars spent in order to not have the slightest amount of quality of life improvement, if anything a worse quality of life because you buy an EV that you cannot use 24/7/365 whereas you can an ICE car for much less .
As soon as something kinda elegant and hopeful as far as collective quality of life improvement is concerned (AI/ChatGPT) came around.....the whole green/EV revolution rightfully went out the window
If Musk was this genius you guys make him to be at 50 and with all the capital he burned he should have at least one company that if you disappeared the world would look drastically different, like if you disappeared Microsoft or Apple or Exxon or Aramco or Amazon or IBM....the world would come to a screeching halt.
Disappear one of Musk companies and everything would be the same as he's always involved in these sort of aspirational companies which have this great vision always 5 years into the future that never materialize into anything tangible or that improves the quality of life like the company I mentioned earlier
well Tesla did jump start the EV revolution not life altering but is pretty important. IF SpaceX gets spaceship right that will be a huge leap forward.
Sir, your comment appears to qualify as "moving the goal post". TSLA never delivered a single inexpensive electric vehicle, and just last week abandoned all high-end efforts (S/X/CT discontinued). All TSLA manufactures now are overpriced "meh" transport boxes. Yes, TSLA was early, and now they are far, far behind the competition.
Can we evaluate based on the stated goals, or why does the criteria keep shifting?
Tesla's goal was to accelerate the world's transition to sustainable energy by building a comprehensive ecosystem of electric vehicles (EVs), solar generation, and battery storage.
I'm amazed at this kind of thinking. I get it, obviously, and it's not uncommon, but still.
Elon Musk has already revolutionized three industries:
1. EVs: Before Tesla, no one thought electric cars could be a mass-market product. And even today, the Model 3 and Model Y are at the top of almost all sales lists.
2. Orbital Launch: No one expected Space X to succeed. What does a software guy know about real engineering? But today, re-usable rockets are the way of the future, and Space X is at least 5 to 10 years ahead of any other company.
3. Satellite Communications: Every single major military power is trying to deploy their own version of Starlink. Before Starlink, 50 satellites was considered a big constellation. Starlink has 8,000 satellites and they are literally launching hundreds every month.
I know it's impossible to prove a counter-factual, but I'm convinced that none of these three would have happened without Elon. No other Western car company has (even now) produced a profitable EV. No other space company has prices as low as Space X. No one even has the capability to build a Starlink competitor (not yet at least). Without Elon pushing these projects, they simply would not have happened or would have happened decades later (after China or someone else beat us to it).
Even his not-yet-successful projects are far beyond most other companies:
Boring Company has actually built tunnels and passengers are actually riding it. No one else is even trying.
Neuralink has actually helped patients.
Tesla FSD actually does work (I use it all the time), and even if Waymo is ahead, Tesla is easily in second place.
I 100% get the hatred for Elon Musk. His political positions are absolutely worth criticizing and I cringe most of the time he tweets. But to deny his business and engineering ability is just motivated reasoning.
Such illusions are ultimately self-defeating. The more opposed one is to Elon Musk (in business or politics) the more important it is to see his capabilities clearly.
> Boring Company has actually built tunnels and passengers are actually riding it. No one else is even trying.
Boring Company bought an existing tunnel boring machine (TBM), and used it to dig a car tunnel. Their only “innovation” in terms of any cost savings is to dig smaller tunnels - which we already knew could be done (tunnel cost grows with diameter), and which we don’t do for good reasons (capacity, emergency egress).
The branding and marketing exercise was excellent though.
Lol this is why you aren't a VC. Even if every single Musk venture failed other than SpaceX, the investments would have paid off wildly well. You aim for the tails not the median.
In a way, its kind of cool to see how robber barons work in real time in our generation. Its also insanely depressing as they will systematically enshittify and extract as much wealth from society as is possible.
I don't actually think the Robber Barons in the 1920s had people going out of their way to defend them and insist they had special knowledge.
The New Deal happened with massive popular support because people did not like the Barons, and wanted to stop them and actually have a life worth living.
The Robber Barons weren't in the 1920s; that refers to industrial age monopolists (e.g. rail/oil), and culminated in the Sherman Antitrust (i.e. 1800s).
Broadly, your point is still valid, though. Just a mild inaccuracy between the Gilded Age and the roaring 20s.
Not sure if this is stock manipulation trying to push the bubble a little further in a way that doesn't require showing something of substance immediately, or if Elon's having another manic episode after doing too many or not enough drugs, but who didn't see this coming?
I'm sure next week he'll have SpaceX be bought by The Boring Company, sell that to Tesla, then rename all the companies as "X".
Also, whatever happened to his plan to turn twitter into a financial services company?
> "The only logical solution therefore is to transport these resource-intensive efforts to a location with vast power and space. I mean, space is called 'space' for a reason. [crying laughing emoji]"
This is all the reasoning provided. It is quite sad how a company I admired so much has become embroiled in financial doohickery.
> Starship will be capable of landing massive amounts of cargo on the Moon […] to establish a permanent presence and take advantage of lunar resources to manufacture satellites and deploy them further into space.
The trouble with strip mining the moon is that it is a pristine international geological park where one side is permanently visible from Earth*. In terms of park visits it’s been seen by pretty much every human that ever existed. Take that, Yosemite. The far side will be banned from exploitation to maintain its unique park status as being almost completely radio silent.
Perhaps the mining will take place behind the ridge line of limbward mountains: technically on the near side but without being visible. Going underground feels like a bit of a stretch.
On the far side, how far does one have to be from the anti-Earth point before one can fire up the WiFi without pissing off the space telescopes?
Who will even regulate this stuff? Do we extend the Antarctic treaty for whole-lunar purposes?
*Worst case, a 5km wide strip mine is 10 pixels on a DSLR photo, but that’s still too much for some.
Elon Musk’s AI startup, xAI (x.ai = grok), officially acquired X Corp. (the parent company of the social media platform X or x.com, formerly Twitter/x.com) in an all-stock deal. Both now operate under a unified holding entity, frequently referred to in corporate filings as X.AI Holdings Corp. (or simply xAI). Now SpaceX has moved to acquire or merge with xAI. This effectively brings the social media data from x.com, the AI development of x.ai, and the satellite infrastructure of Starlink/SpaceX under one "super-conglomerate" roof.
> The basic math is that launching a million tons per year of satellites generating 100 kW of compute power per ton would add 100 gigawatts of AI compute capacity annually, with no ongoing operational or maintenance needs. Ultimately, there is a path to launching 1 TW/year from Earth.
> My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space.
This is so obviously false. For one thing, in what fantasy world would the ongoing operational and maintenance needs be 0?
You operate them like Microsoft's submerged data center project: you don't do maintenance, whatever fails fails. You start with enough redundancy in critical components like power and networking and accept that compute resources will slowly decrease as nodes fail
No operational needs is obviously ... simplified. You still need to manage downlink capacity, station keeping, collision avoidance, etc. But for a large constellation the per-satellite cost of that would be pretty small.
How do you make a small fortune? Start with a big one.
The thing being called obvious here is that the maintenance you have to do on earth is vastly cheaper than the overspeccing you need to do in space (otherwise we would overspec on earth). That's before even considering the harsh radiation environment and the incredible cost to put even a single pound into low earth orbit.
If you think the primary source of electricity is solar (which clearly Musk does), then space increases the amount of compute per solar cell by ~5x, and eliminates the relatively large battery required for 24/7 operation. The thermal radiators and radiation effects are manageable.
The basic idea of putting compute in space to avoid inefficient power beaming goes back to NASA in the 60s, but the problem was always the high cost to orbit. Clearly Musk expects Starship will change that.
ISS cooling is 16KW dissipation. So like 16 H200. Now imagine you want to cool 100k instead of 16.
And all this before we talk about radiation, connectivity (good luck with 100gbps rack-to-rack we have on earth), and what have you.
—
Sometimes I think all this space datacenters talk is just a PR to hush those sad folks that happen to live near the (future) datacenter: “don’t worry, it’s temporary”
> ROSA is 20 percent lighter (with a mass of 325 kg (717 lb))[3] and one-fourth the volume of rigid panel arrays with the same performance.
And that’s not the current cutting edge in solar panels either. A company can take more risks with technology choices and iterate faster (get current state-of-the-art solar to be usable in space).
The bet they’re making is on their own engineering progress, like they did with rockets, not on sticking together pieces used on the ISS today.
How much maintenance do you need? Lets say you have hardware whose useful lifespan due to obsolescence is 5 years, and in 4, the satellite will crash into the atmosphere anyways.
Let's say given component failure rates, you can expect for 20% of the GPUs to fail in that time. I'd say that's acceptable.
A lot. As someone that has been responsible for trainings with up to 10K GPUs, things fail all the time. By all the time I don't mean every few weeks, I mean daily.
From disk failings, to GPU overheating, to infiniband optical connectors not being correctly fastened and disconnecting randomly, we have to send people to manually fix/debug things in the datacenter all the time.
If one GPU fails, you essentially lose the entire node (so 8 GPUs), so if your strategy is to just turn off whatever fails forever and not deal with it, it's gonna get very expensive very fast.
And thats in an environment where temperature is very well controlled and where you don't have to put your entire cluster through 4 Gs and insane vibrations during take off.
The idea here is that the economics of launch are changing with Starship such that the "incredible cost" and "overspeccing" of space will become much less relevant. There's a world where, because the cost per kg is so low, a data center satellite's compute payload is just the same hardware you'd put in a terrestrial rack, and the satellite bus itself is mass-produced to not-particularly-challenging specs. And they don't have to last 30 years, just 4-ish, when the computer is ready for retirement anyway.
Will that come to be? I'm skeptical, especially within the next several years. Starship would have to perform perfectly, and a lot of other assumptions hold, to make this make sense. But that's the idea.
If you ramp up the economies of scale to make those things - radiation protection and cost per pound - the calculus changes. It's supposed to synergize with Starship, and immediately take advantage of the reduced cost per pound.
If the cost per pound, power, regulatory burden, networking, and radiation shielding can be gamed out, as well as the thousand other technically difficult and probably expensive problems that can crop up, they have to sum to less than the effective cost of running that same datacenter here on earth. It's interesting that it doesn't play into Jevon's paradox the way it might otherwise - there's a reduction in power consumption planetside, if compute gets moved to space, but no equivalent expansion since the resource isn't transferable.
I think some sort of space junk recycling would be necessary, especially at the terawatt scale being proposed - at some point vaporizing a bunch of arbitrary high temperature chemistry in the upper atmosphere isn't likely to be conducive to human well-being. Copper and aluminum and gold and so on are also probably worth recovering over allowing to be vaporized. With that much infrastructure in space, you start looking at recycling, manufacturing, collection in order to do cost reductions, so maybe part of the intent is to push into off-planet manufacturing and resource logistics?
The whole thing's fascinating - if it works, that's a lot of compute. If it doesn't work, that's a lot of very expensive compute and shooting stars.
Or, just saying, be critical of ideas and think them through, and take in what experts say about it, and determine for yourself what's up. If a bunch of people who usually seem to know what they're talking about think there's a legitimate shot at something you, as a fellow armchair analyst, think is completely impractical, it makes sense to go and see if maybe they know something you don't.
In this case, it's all about Starship ramping up to such a scale that the cost per pound to orbit drops sufficiently for everything else to make sense - from the people who think the numbers can work, that means somewhere between $20 and $80 per pound, currently at $1300-1400 per pound with Falcon 9. Starship at scale would have to enable at least 2 full orders of magnitude decrease in price to make space compute viable.
If Starship realistically gets into the $90/lb or lower range, space compute makes sense; things like shielding and the rest become pragmatic engineering problems that can be solved. If the cost goes above $100 or so, it doesn't matter how the rest of the considerations play out, you're launching at a loss. That still might warrant government, military, and research applications for space based datacenters, especially in developing the practical engineering, but Starship needs to work, and there needs to be a ton of them for the datacenter-in-space idea to work out.
Or, just saying, we should eat babies because they are abundant and full of healthy nutrition for adult humans. [1]
Just because an idea has some factors in its favor (Space-based datacenter: 100% uptime solar, no permitting problems [2]) doesn't mean it isn't ridiculous on its face. We're in an AI bubble, with silly money flowing like crazy and looking for something, anything to invest it. That, and circular investments to keep the bubble going. Unfortunately this gives validation to stupid ideas, it's one of the hallmarks of bubbles. We've seen this before.
The only things that space-based anything have advantages on are long-distance communication and observation, neither of which datacenters benefit from.
The simple fact is that anything that can be done in a space-based datacenter can be done cheaper on Earth.
Note how Musk cleverly doesn't claim that not doing maintenance drives down costs.
Nothing in there is a lie, but any substance is at best implied. Yes, 1,000,000 tons/year * 100kW/ton is 100GW. Yes, there would be no maintenance and negligible operational cost. Yes, there is some path to launching 1TW/year (whether that path is realistic isn't mentioned, neither what a realistic timeline would be). And then without providing any rationale Elon states his estimate that the cheapest way to do AI compute will be in space in a couple years. Elon is famously bad at estimating, so we can also assume that this is his honest belief. That makes a chain of obviously true statements (or close to true, in the case of operating costs), but none of them actually tell us that this will be cheap or economically attractive. And all of them are complete non-sequiturs.
Unless I missed something the Microsoft underwater data center was basically a publicity stunt.
Anyone who thinks it makes sense to blast data centers into space has never seen how big and heavy they are, or thought about their immense power consumption, much less the challenge of radiating away that much waste heat into space.
I was listening to a Darknet Diaries episode where Maxie Reynolds seems to make it work: https://subseacloud.com/ I don't know how profitable they are, and I doubt this is scalable enough, but it can work as a business.
Well the thing is that it seemed to have been successful beyond all expectations despite being that? They had fewer failures due to the controlled atmosphere, great cooling that took no extra power, and low latency due to being close to offshore backbones. And I presume you don't really need to pay for the land you're using cause it's not really on land. Can one buy water?
Space is pretty ridicolous, but underwater might genuinely be a good fit in certain areas.
Hot saltwater is the worst substance on earth, excepting, maybe, hydrofluoric acid. You really don't want to cool things with ocean water over an extended period of time. And filtering/purifying it takes vast amounts of power (e.g. reverse osmosis).
An 8 GPU B200 cluster goes for about $500k right now. You'd need to put thousands of those into space to mimic a ground-based data center. And the launch costs are best case around 10x the cost of the cluster itself.
Letting them burn up in the atmosphere every time there's an issue does not sound sustainable.
A Falcon Heavy takes about 63 tons to LEO, at a cost of about $1,500 per kg.
A server with 4x H200s and some RAM and CPU costs about $200k, and weighs about 60kg, with all the cooling gear and thick metal. As is, it would cost $90k to get to LEO, half of the cost of the hardware itself.
I suppose that an orbit-ready server is going to cost more, and weigh less.
The water that serves as the coolant will weigh a lot though, but it can double as a radiation shield, and partly as reaction mass for orbital correction and deorbiting.
Power draw is max 10.2 kW but average draw would be 60-70% of that. let's call it 6kW.
It is possible to obtain orbits that get 24/7 sunlight - but that is not simple. And my understanding is it's more expensive to maintain those orbits than it would be to have stored battery power for shadow periods.
Average blackout period is 30-45 minutes. So you'd need at least 6 kWh of storage to avoid draining the batteries to 0. But battery degradation is a thing. So 6 kWh is probably the absolute floor. That's in the range of 50-70 kg for off-the-shelf batteries.
You'd need at least double the solar panel capacity of the battery capacity, because solar panels degrade over time and will need to charge the batteries in addition to powering the gpu's. 12 kW solar panels would be the absolute floor. A panel system of that size is 600-800 kg.
These are conservative estimates I think. And I haven't factored in the weight of radiators, heat and radiation shielding, thermal loops, or anything else that a cluster in space might need. And the weight is already over 785 kg.
Using the $1,500 per kg, we're approaching $1.2 million.
Again, this is a conservative estimate and without accounting for most of the weight (radiators) because I'm too lazy to finish the napkin math.
I think we're on the same page. Lifting the actual computing devices would be not that expensive, compared to lifting a lot of other related mass, principally the cooling systems, and the solar panels.
The solar panels used in space are really lightweight, about 2 kg / m² [1], it's like ten times lighter weight than terrestrial panels. Still they need load-bearing scaffolding, and electrical conductors to actually collect the hundreds of kilowatts.
Are launch costs really 10x!? Could I get a source for that?
In the back on my head this all seemed astronomically far-fetched, but 5.5 million to get 8 GPUs in space... wild. That isn't even a single TB of VRAM.
Are you maybe factoring in the cost to powering them in space in that 5 million?
You mean you operate them like Microsoft's failed submerged data center project [1]. When pointing at validating past examples you are generally supposed to point at successes.
> By 2024, Project Natick had been inactive for several years, though it was referenced in media as though it was ongoing. That year, Microsoft confirmed that the project was inactive and that it had no servers underwater.
I wouldn't exactly call this a success, for that matter.
But if we’re going down that line of thinking then it’s a poor comparison. I could open a data centre on the ground and use the same principle of zero maintenance, and it would be way cheaper and way more powerful.
This will totally work since we have an unlimited amount of rare earth elements we can just ship off into space never to see again. Infinite raw materials + infinite power equals infinite AI!!!
My understanding was that access to very large body of cold water was a core feature for the project. The water was to be used for cooling relatively efficiently or cheaply.
The thing is: at the end of the day, SpaceX takes the "impossible" and makes it "late".
People are going to Tory Bruno the space datacenters until one day their Claude agent swarm's gonna run in space and they'll be wondering "how did we get here"?
The thing is: at the end of the day, making absolute statements about the inevitability of future success is a fool’s errand.
Musk has a documented history of failing to deliver on promises, timescale or no. So it’s best to engage in some actual critical thinking about the claims he is making.
There's clearly rhetorical hyperbole happening there. But assuming that thermal rejection is good in space, & launch costs continue falling, as earth-based data centers become power/grid-constrained, there is a viable path for space power gen.
The craziest part of those statements is "100 kW per ton." IDK what math he is doing there or future assumptions, but today we can't even sniff at 10 kW per ton. iROSA [1] on the ISS is about 0.150 kW per ton.
Not to be an Elon defender, but can you back up your 0.15/ton? My own searching puts ROSA orders of magnitude higher. Each array is 600kg (0.6t) and puts out 20kw (https://space.skyrocket.de/doc_sdat/irosa-1.htm) which makes 20/0.6 = 33.333 kw/ton
Hey all good. My advice, not that you asked for it, is to put the math in the comment. Even as a footnote. I've found myself backtracking a lot of math comments after I stare at it in the text box for a few seconds.
Maybe I'm just out of the loop, but is solar substantially more efficient in space? I assume the satellites won't orbit in a way that follows the sun. And presumably the arrays of panels they can attach to a satellite don't exceed the size of the panels you could slap on and around a data center (at least without being insanely expensive).
1. solar is very efficient at generating energy, no moving parts, simple physics etc.
2. in space you don't deal with weather or daylight cycle, you can just point your panels at the sun and generate very stable energy, no batteries required
3. environmental factors are simpler, no earthquakes, security, weather. Main problem here is radiation
In theory its a very elegant way to convert energy to compute.
2 is wrong. At a Lagrange point you can do this. Not in low earth orbit - in LEO sunset is every 60 minutes or so, and you spend the next 60 minutes in darkness.
Satellites are heavily reliant on either batteries or being robust to reboots, because they actually do not get stable power - it's much more dynamic (just more predictable too since no weather).
The US mandates by law that we grow a fuck ton of corn to mix 10% ethanol into gasoline.
If you replaced just those cornfields with solar/wind, they would power the entire USA and a 100% electric vehicle fleet. That includes the fact that they are in the corn belt with less than ideal
sun conditions.
We aren’t even talking about any farmland that produces actual food or necessary goods, just ethanol as a farm subsidy program.
The US is already horrendously bad at land use. There’s plenty of land. There’s plenty of ability to build more grid capacity.
Solar in space is a very different energy source in terms of required infrastructure. You don't need batteries, the efficiency is much higher, cooling scales with surface area (radiative cooling doesn't work as well through an atmosphere vs. vacuum), no weather/day cycles. Its a very elegant idea if someone can get it working.
The panels suffer radiation damage they don't suffer on Earth. If this is e.g. the same altitude orbits as Starlink, then the satellites they're attached to burn up after around tenth of their ground-rated lifetimes. If they're a little higher, then they're in the Van Allen belts and have a much higher radiation dose. If they're a lot higher, the energy cost to launch is way more.
If you could build any of this on the moon, that would be great; right now, I've heard of no detailed plans to do more with moon rock than use it as aggregate for something else, which means everyone is about as far from making either a PV or compute factory out of moon rock as the residents of North Sentinel Island are.
OK, perhaps that's a little unfair, we do actually know what the moon is made of and they don't, but it's a really big research project just to figure out how to make anything there right now, let alone making a factory that could make them cost-competitive with launching from Earth despite the huge cost of launching from Earth.
> The panels suffer radiation damage they don't suffer on Earth.
I don't think this is true, Starlink satellites have an orbital lifetime of 5-7 years, and GPUs themselves are much more sensitive than solar panels for rad damage. I'd guess the limiting factor is GPU lifetime, so as long as your energy savings outpace the slightly faster gpu depreciation (maybe from 5 -> 3 years) plus cost of launch, it would be economical.
I've said this elsewhere, but based on my envelope math, the cost of launch is the main bottleneck and I think considerably more difficult to solve than any of the other negatives. Even shielding from radiation is a weight issue. Unfortunately all the comments here on HN are focused on the wrong, irrelevant issues like talking about convection in space.
Elon is a pathological liar and it’s crazy that he still gets sanewashed after all he’s done. It’s insanity that he hasn’t been kicked out of leading his companies, and it’s also insanity that he hasn’t been prosecuted by the SEC.
You’ve spent too much life force trying to even understand the liar’s fake logic.
Let’s start right here: there is no such thing as becoming power/grid constrained on earth. If you replaced just the cornfields that the United States uses just to grow corn for ethanol in gasoline just in the corn belt, you could power the entire country with solar+batteries+wind. Easily, and cheaply.
If you don’t even believe that solar+batteries are cheap (they are), fine, choose your choice of power plant. Nuclear works fine.
The truth is, xAI combining with SpaceX is almost certainly corrupt financial engineering. SpaceX as a government contractor and that means Elon’s pal Trump can now siphon money into xAI via the federal government.
This is par for the course for an Elon-associated endeavor but it's been leaking out into the broader tech sector; make ludicrous claims and promises and somehow investors just throw money at you. FSD has been around the corner for over a decade, martian colonization will be here by the end of the decade for the past 20 years and General SuperAI will be here in a few years for the past few years.
Currently, just a cursory google search shows $1500-3000 per kilogram to put something into low earth orbit. Lets take the low bound because of efficiencies of scale. So $1500.
A million tons will cost $1500x1000x1000000= 1,500,000,000,000. That is one and a half TRILLION dollars per year. That is only the lift costs, it does not take into account the cost of manufacturing the actual space data centers. Who is going to pay this?
Yes, and as we know Starship will be doing regular commercial launches starting in 2020, maybe 2021.
We're getting close to having the time for Starship's delays to be the same as the actual time for the Saturn 5 to go from plans to manned launches (Jan 1962-Dec 1968).
It’s hard to estimate what Starship’s actual costs will be when it isn’t fully operational. I am finding estimates of $100 to $200 per kilogram and even as low as $10 per kilogram.
Let’s say the costs in 5 years do get as low as $15 per kilogram or about 2 orders of magnitude improvement in launch prices. That means a 200-ton payload Starship would cost $3,000 to launch.
Do you honestly believe that? The world’s largest rocket cost a total of $3,000 to launch?
> Let’s say the costs in 5 years do get as low as $15 per kilogram or about 2 orders of magnitude improvement in launch prices. That means a 200-ton payload Starship would cost $3,000 to launch. Do you honestly believe that? The world’s largest rocket cost a total of $3,000 to launch?
You have missed three zeroes in this calculation ;)
15 per kg for a 200-ton payload is about 3 million$. That seems achievable, given that propellant costs are about 1-1.5 million.
"it'll never work" is quite black and white while "failure" is a lot more of a grey area. Will it actually launch? Sure, we've seen it. Will it actually hit the reliability as sold? Will it have as fast of turnaround time to reach launch timing goals? Can it actually launch as much payload as promised? Will the economics actually shake out as intended?
Did the Cybertruck "never work"? Obviously not, they're on the streets. Was it a <$40k truck with >250mi range? No.
Did FSD "never work"? Obviously not, tons of people drive many, many miles without touching the wheel. Does Tesla feel confident in it enough to not require safety operators to follow it on robotaxi trips? No. Does Tesla trust it enough to operate in the Las Vegas Loop? No. Has Tesla managed to get any state to allow it to operate truly autonomously? No.
Look, I hope Starship does work as advertised. Its cool stuff. But I don't see it as a given that it will. And given by the track record of the guy who promised it, it gives even less confidence. I'm sad there's less competition in this space. We have so many billionaires out there and yet so few out there actually willing push envelopes.
One reliable method of pushing envelopes, attracting investment and hiring smart people is to get excited about unrealistic timelines.
The best case is you meed the unrealistic timeline, the average case outcome is you solve the problem but it is delayed several years. And the worst case is it fails and investors lose some money.
If you try to hire people but your message is: we want to reduce the cost of access to space by 20% in thirty years, you are going to get approximately zero competent engineers, and a whole lot of coasters.
And no investors, so you'll be dependent on the government anyway. Depending on the government is great until people you do not agree with or are generally anti science, are in power. I assume this part should not need an example nowadays?
> One reliable method of pushing envelopes, attracting investment and hiring smart people is to get excited about unrealistic timelines.
Its also a good way to shred morale and investor confidence when you're a decade past your timelines or continue to fail on actually delivering on past promises.
That launch cost is remarkably cheap to someone that's handled a $1.5million dollar 5U server filled with GPUs and RAM that weighs under 100kg.
Obviously the solar and cooling for the above would both weigh and cost a ton but... It's feels surprisingly close to being within an order of magnitude of current costs when you ballpark it?
Like i don't think it's actually viable, it's just a little shocking that the idea isn't as far out of line as i expected.
I remember reading somewhere that satellites are extra expensive for 2 reasons:
- launch costs are so high that doing exotic bespoke engineering might be worth it if it can shave off a few pounds
- once again because launches are expensive and rare, you cannot afford to make mistakes, so everything has to work perfectly
If you are willing to launch to lower orbits, and your launch vehicle is cheap, you are building in bulk, then you can compromise on engineering and accept a few broken sats
Undergrads afaik even high schoolers have built cubesats out of aluminum extrusions, hobbyist solar panels, and a tape measure as an antenna. These things probably dont do that much, but they are up there and they do work.
They are also expensive because there are unique challenges to making reliable spacecraft. E.g. cosmic rays and microgravity absolutely wreck electronics. Those undergrad cube sats are lucky to last more than a few months in the relative calm of low inclination, low altitude earth orbit. They would die on their first pass in a sun synchronous polar orbit.
Here is my main question: Musk is on record as being concerned about runaway "evil AI." I used to write that off as sci-fi thinking. For one thing, just unplug it.
So, let's accept that Musk's concern of evil runaway AI is a real problem. In that case, is there anything more concerning than a distributed solar powered orbital platform for AI inference?
Aside from anything about Elon Musk, here’s an interesting video response to the “just unplug it” argument on the Compuerphile channel: https://youtu.be/3TYT1QfdfsM
Ha, I figured that might be the video prior to clicking it. I am a long time fan.
Agreed, when I wrote "just unplug it," this counterargument was present in my mind, but nobody likes a wall of text.
However, my original point was that a distributed solar powered orbital inference platform is even worse! Think about how hard it would be to practically take out Starlink... it's really hard.
Now.. >1M nodes of a neural net in the sky? Why would someone who lives as a god, the richest man in the world, the only person capable of doing this thanks to his control of SpaceX... do the literal worst thing possible?
tone: I don't really understand orbital mechanics, but I do understand geopolitics a bit.
1. China is very concerned about Starlink-like constellations. They want their own, but mostly they want to be able to destroy competitors. That is really hard.
2. Many countries have single ASAT capabilities. Where one projectile can hit one satellite. However, this is basically shoot a bullet, with a bullet, on different trajectories.
3. > Sure, it'd take orbital launch capabilities to lift ... how many bags of metal scrap and explosives?
If I understand orbital mechanics... those clouds of chaff would need to oppose the same orbit, otherwise it is a gentle approach. In the non-aligned orbit, it's another bullet hitting a bullet scenarios as in 2, but with a birdshot shotgun.
My entire point is that constellations in LEO take hundreds of Falcon 9's worth of mass to orbit and delta-v to destroy them, as in-orbit grenades which approach gently. This IS REALLY HARD, as far as mass to orbit, all at once! If you blow up some group of Starlink, that chaff cloud will just keep in orbit on the same axis. It will not keep blowing up other Starlinks.
The gentle grenade approach was possibly tested by the CCP here:
> tone: I don't really understand orbital mechanics, but I do understand geopolitics a bit.
Thanks for the clarification, I guess that explains this (from you):
> Think about how hard it would be to practically take out Starlink.
and this:
> My entire point is that constellations in GEO
which you've now corrected.
Moving on:
> My entire point is that constellations in LEO take hundreds of Falcon 9's worth of mass to orbit and delta-v to destroy them, as in-orbit grenades which approach gently. This IS REALLY HARD
So let's not do that .. how hard is it to render the entire LEO zone a shit show with contra wise clouds of frag that cause cascading failures?
Forget the geopolitics of China et al. .. LEO launch capabilities are spreading about the globe, it's not just major world powers that pose a threat here.
Ok... so, let's reset, please. I bet that we have very similar intentions, and yet on internet forums, we have perfected the art of users speaking past each other.
Just to get on the same page here. My arugument is that prior to Elon Musk, the only human capable of launching >1M distributed solar powered inference nodes, if one accepts runaway AGI/ASI as a threat... prior to that we had a few hundred terrestrial AI inference mega-data centers. Most of them had easily disrupted power supplies by one dude with a Sawzall.
Now, we are moving to a paradigm where the power supply is the sun, the orbital plane gives the nodes power 24/7, and the dude with the Sawzall needs to buy >10,000x (not sure of the the multiple here) the Sawzalls, and also give them escape velocity.
Can we not agree that this is a much more difficult problem to "just unplug it," than it was when the potentially troublesome inference was terrestrial?
My up thread commentary was not meant as real snark at all. I was attempting to be genuine.
However, I think it did accomplish my goal. I bet that we could now have a beer/tea, and laugh together.
If you are ever near Wroclaw, Prague, Leipzig/Dresden, or Seattle, please email my username at the the big G. I would happily meet you at the nearest lovely hotel bar. HN mini meetup. I can only imagine the stories that we might exchange.
Look, I'm Australian, I enjoy a bit of banter. I stripped the personal info from my comment above; I was happy to share with you, reluctant to leave it as was.
I was a frequent Toronto visitor, for the TSX, back when we ran a minerals intelligence service before passing that onto Standard&Poor.
You're on the list, however my movements are constrained for now, my father's a feriously active nonagenarian which is keeping me with one foot nailed to the ground here for now.
Also, thank you for the reminder that I need to get my ass back to Seattle to be with remaining parent, while I still can. I have been a jackass about that.
What, creating a huge patchwork of self sufficient AIs, forming their own sky based net, seems bad to you, considering the whole torment nexus/Sky Net connotations? It's not like he's planning to attach it to his giant humanoid robot program. Oh. Ohhhhh. Oh no.
It is already more expensive to performance maintenance on SOCs than it is to replace them. Remember, these machines are not for serving a database, there are practically no storage needs (and storage is the component that fails most often.)
Given that, the main challenge is cooling, I assume that will be figured out before yeeting 100 billion $ of computers into space. Plenty of smart people work at these companies.
Do smart people work at Boring Company? Do smart people work on FSD at Tesla? What about the HyperLoop? It is possible for smart people to make technical achievements without the overall project being particularly successful.
I meant it specifically for figuring out cooling computers in space.
I am pretty sure this is going to be a solvable problem if this is the bottleneck to achieve data centers in space, given that newer chips are much more tolerant to high temperatures.
Any estimate by Elon musk, you need to add or substract a zero to/from the end. Here, I'll fix it for you.
> The basic math is that launching a 100,000 tons per year of satellites generating 10 kW of compute power per ton would add 1 gigawatt of AI compute capacity annually, with no ongoing operational or maintenance needs. Ultimately, there is a path to launching 0.01 TW/year from Earth.
> My estimate is that within 20 to 30 years, the lowest cost way to generate AI compute will be in space.
One of the biggest but most pointless questions I have about our current moment in history is whether the people in power actually believe the stuff they say or are lying. Ultimately I don't think the answer really matters, their actions are their actions, but there is just so much that is said by people like Musk that strains credulity to the point that it indicates either they're total idiots or they think the rest of us are total idiots and I'm genuinely curious which of those is more true.
We’re at a point where propaganda is so much more powerful than reality that the people in power literally can’t tell the difference. When your source of ethics is the stock price, little details like physical impossibility stop seeming relevant.
You put it so succinctly and perfectly that I'll have to favorite your comment. Totally agree. The physical world has become little more than noise for people like Musk. I wonder whether the correction will be a slow market dip, a full collapse, or somehow whether he makes it out like a bandit. Baudrillard is, once again, uncomfortably accurate in his diagnosis.
But more importantly, there is no heat dissipation in space. There’s no atmosphere to cool you, no water you can put heat into. Just an empty void. You can radiate a little, but the sun alone is enough to cook you, without you having a rack of GPUs inside your satellite.
It’s completely delusional to think you could operate a data centre in a void with nowhere to put the heat.
Never mind operational and maintenance costs. In what fantasy world is it cheaper to put a computer in orbit than in a building on the ground? I don't care how reusable and maintenance-free Starship gets, there's no way even absurdly cheap launches are cheaper than a building.
The whole thing makes no sense. What's the advantage of putting AI compute in space? What's even one advantage? There are none. Cooling is harder. Power is harder. Radiation is worse. Maintenance is impossible.
The only reason you'd ever put anything in orbit, aside from rare cases where you need zero-gee, is because you need it to be high up for some reason. Maybe you need it to be above the atmosphere (telescopes), or maybe you need a wide view of the earth (communications satellites), but it's all about the position, and you put up with a lot of downsides for it.
I feel like either I'm taking crazy pills, or all these people talking about AI in space are taking crazy pills. And I don't think it's me.
The most generous interpretation is that the "AI in space" nonsense is a cover for putting limited AI in space for StarShield (military version of StarLink), which is essentially the "Golden Dome".
It might be possible to scam the Pentagon with some talk about AI and killer satellites that take down ICBMs.
You do realize that “space-grade” involves process changes that intrinsically incur orders of magnitude efficiency losses? Larger process sizes, worse performing materials. It’s not just a design thing you can throw money at.
Only if that bitflip happens somewhere in your actual data, vs. some GPU pipeline register that then locks up the entire system until a power cycle. Or causes a wrong address to be fetched. Or causes other nasty silent errors. Or...
Try doing fault injection on a chip some time. You'll see it's significantly easier to cause a crash / reset / hang than to just flip data bits.
'rad-triggered bit flips don't matter with AI' is a lie spoken by people who have obviously never done any digital design in their life.
The problem is not solved and the techniques they use to deal with it run directly contrary to maximizing compute, because that's not historically something they have remotely cared about.
Politics and finances aside, I wonder how "sending megatons of mass" into space is more ecological than building power plants needed for data centers on earth? Not only all the fuel that you'd need to burn, but also the fact that this material probably can't be recycled since it will burn on reentry.
Just a neat bit of financial engineering. You can tell because Elon picked SpaceX instead of Tesla – which would have actually made sense at some level (Optimus Robots + AI). But Tesla is public and so he'd need to follow laws and reporting requirements.
You can tell it's just financial engineering because in the entire press release xAI is only mentioned in the first two sentences. Everything after that is Elon talking about space data centers to distract from the actual topic. Which seems to be working
Tesla is still very profitable, as is SpaceX I assume. Twitter/X has been a $44 billion dollar failure, and xAI is a vanity project so Musk can go around saying he is a player in the AI space. Investors in both X and xAI need to be bailed out, hence this announcement.
In the world where it makes $8-10B in profit on $90-95 billion in revenue every year. Whatever price investors choose to trade the stock at is irrelevant to those numbers.
The $44B Twitter/X buyout was not a failure. For example Fidelity has its $19M investment in the buyout - now xAI common shares - marked at $62M (up over 3X) as of 12/31/25. It was certainly valued even higher on 1/31/26 after xAI had an oversubcribed fund raise in January. All before this merger announcement.
The fact that it had to be successively bailed out by xAI (which itself was funded by Tesla) and now SpaceX shareholders is exactly what makes the acquisition a failure.
He spent other people's money (or maybe even imaginary money) he couldn't have used for himself (since selling off major stakes in your company is a big nono)
Do you genuinely not think that "Elon" (xAI) is player in the AI space?
You don't have to think they have the best models of course, but they are clearly a very significant, and some might argue, leading player in the AI race.
xAI’s models are really not pioneering at all. They weren’t the first to do MoE. Not the first to do open weighting, not the first to have memory or multi-modal vision.
So no, I wouldn’t say Elon is a major player in the AI space. People use his models because they are cheap and are willing to undress people’s photos.
saying they aren't pioneering is very different than saying they aren't a major player in the space. There're only like 5-7 players with a foundational model that they can serve at scale. xAI is one of them
I suspect SpaceX will acquire Tesla at some point. It’s the most profitable of these companies. So basically SpaceX employees and shareholders are covering up for the failing Tesla business and the already-failed xAI business.
Let’s not forget, xAI is the parent of Twitter/X (the social network). So now, taxpayers are paying to keep Twitter/X alive. After all, it is taxpayer money going to the contracts the government gives SpaceX for launches. Nice way to subsidize what is effectively a one sided campaign machine for the GOP and far right.
> I suspect SpaceX will acquire Tesla at some point.
I think that is also likely, unless Tesla can stage a major turnaround, it is going to be beaten by Chinese competitors nearly everywhere that they are allowed (which is everywhere but the USA.)
This was my immediate thought as well. A great time to ask yourself — why am I literally paying for any of this? At best I literally don't use any of these services, at worst they are actively used against me.
I get what you're saying, but that taxpayer money is paying for the launch services at a very competitive rate (possibly the cheapest of all available options), not a subsidy scheme.
I guess the difference is Tesla is a public company, so requires more paperwork. SpaceX isn't public yet, but will be soon, meaning it will have a cash infusion.
This is fairly naive, Elon isn't the only investor in SpaceX.
My guess is "that they did the math" and had an engineering study which convinced them that getting AI datacenters into space will make sense.
It's also not hard to imagine why, the process alone once perfected could be reused for asteroid mining for example, then mirogravity manufacturing, either of which alone would be enormous capital intensive projects. Even if AI dataenters in space are break-even it would be a massive win for SpaceX and leave their competition far behind.
You do not radiate all the heat away from a GPU, a modern GPU can run pretty hot. Also look up how this is getting better for the next generation of GPUs.
Maybe repeat your calculation with updated assumptions?
But even if you were completely right, your argument is that we can't do this tomorrow, yes I agree. Typical technology development cycles are about 5-10 years.
>You do not radiate all the heat away from a GPU, a modern GPU can run pretty hot.
LOL. If you don't radiate the heat the spacecraft just gets indefinitely hotter (until it glows and the heat is forcibly irradiated). It's space, there's no fluid to provide convection.
Jared Vennett (narration):
"In the years that followed, hundreds of bankers and rating agency's executives went to jail. The SEC was completely overhauled, and Congress had no choice but to break up the big banks and regulate the mortgage and derivatives industries."
"Just kidding. Banks took the money the American people gave them, and they used it to pay themselves huge bonuses, and lobby the Congress to kill big reform. And then they blamed immigrants and poor people, and this time even teachers."
Because this move is entirely financial engineering to hide losses just like the roll up of X in to xAI.
None of this has anything to do with business or innovation. Do you not immediately see that? Most of my friends reaction to this news was that this is so obvious it's almost funny (or actually it is funny, since most were laughing as they read the headline).
I'm curious how you could not understand the relevance of the quote unless you were aggressively trying to not understanding it.
I understand it now, after reading the thread. There's a reason for that.
I have not been following the machinations of X very closely. I don't have the corporate structure of Elon's empire in my head, nor do I have the Meta or Alphabet/Google hierarchies in there. I couldn't have told you about the history of xAI beyond that it exists.
So that's plain ignorance of something you consider common knowledge, but I don't, rather than "aggressively trying to not understand it." And that phrase is particularly grating btw.
What are they hiding that wasn't hidden already? Two private companies making a private transaction.. there is no mandatory reporting now nor after this move
SpaceX investors want to cash out, which is why they’re going public. Elon Musk wants to dump his X/xAI bags onto the public markets by merging it with SpaceX.
Essentially means that SpaceX investors are bailing out Elon Musk.
Don't forget that a lot of US mil stuff is launched by SpaceX so in a very real way they are the prime defense contractor in space for the country. If the public offering doesn't work, Unc Sam'll bail them out. Wonder if Trump will want a stake in the company this time.
Because X and xAI are both losing money. X needed cash to operate, so Elon rolled it into xAI to use xAI’s cash to help fund it. xAI is likely burning egregious amounts money, but will have trouble raising more capital. By rolling it into SpaceX he further covers up the financial issues because SpaceX is actually profitable. He can then raise more capital without having to worry (for a while) about how awful the burn is…
I, by and large, have a strong dislike of Musk to put it mildly. The one thing I will give him, and I think this is his real gift, is he’s absolutely brilliant when it comes to raising capital. He has proven to excel at raising capital, and deploying it well, for extremely capital intensive businesses. I do however wonder if the chickens are coming home to roost because both X and xAI are extremely unprofitable.
I think it’s almost inevitable we will see Space X and Tesla merge. The conditions of that merger will, I believe, say a lot about whether this move was brilliant or batshit.
Tesla acquiring solarcity was the same thing over. It did not make sense. Then and it does not make sense now. But the distortion field is so great no one notices.
SolarCity and Tesla made more surface level sense just being in the same general vicinity since they're both fundamentally green energy companies. That made it easy to spin questions about the financials with some CEO-speak about synergy.
However, the way Musk has become less subtle with this tells a story. He got away with these shady financial dealings multiple times so he's now becoming even more brazen and transparent with this behavior. We have gotten to the point in which the spin needed to justify his moves is the physics-defying viability of datacenters in space.
The distortion field will keep growing as long as he keeps getting away with it.
Likely the intended meaning here is that the practicality of space data centers goes against the physical realities of operating in space. The single most prevalent issue with operating anything in space is heat dissipation in that the only method of doing so is via radiation of heat, which is very slow. Meanwhile, the latest Nvidia reference architectures convert such ungodly amounts of power into heat (and occasionally higher share prices) that they call for water cooling and extensive heat-exchange plant.
Even if one got the the economics of launching/connecting GPU racks into space into negligable territory and made great use of the abundent solar energy, the heat generated (and in space retained) by this equipment would prevent running it at 100% utilization as it does in terrestrial facilities.
In addition to each rack worth of equipment you'd need to achieve enough heat sink surface area to match the heat dissipation capabilities of water-cooled systems via radiation alone.
Not physics defying, just economically questionable.
The main benefits to being in space are making solar more reliable and no need to buy real estate or get permits.
Everything else is harder. Cooling is possible but heavy compared to solar, the lifetimes of the computer hardware will probably be lower in space, and will be unserviceable. The launch cost would have to be very low, and the mean time between failure high before I think it would make any economical sense.
It would take a heck of a lot of launches to get a terrestrial datacenter worth of compute, cooling and solar in orbit, and even if you ship redundant parts, it would be hard to get equivalent lifetimes without the ability to have service technicians doing maintenance.
Their viability is what I called physics-defying. Without some drastic changes to our current level of technology, the added costs of putting something in space along with the complexities of powering, cooling, and maintaining it once it's there is just too much to overcome the alternative of just building it on Earth.
Radiative cooling is the only option, and it basically sucks vs any option you could use on earth.
Second, ai chips have a fixed economic life beyond which you want to replace them with better chips because the cost of running them starts to outpaxe the profit they can generate. This is probably like 2-3 years but the math of doing this in space may be very different. But you can't upgrade space based data centers nearly as easily as a terrestrial data center.
But that is also just an assumption isn't it? Could this not also be related to the fact that they plan to launch a ton of servers into the sky to run in space and power AI? It would mean that their AI product would become heavily based on the services provided by SpaceX via launching all this.
But regardless, I think quotes like these should have some commentary around them as it helps create a discussion around whatever point they might be trying to make rather than having to make assumptions.
That's a very disingenuous argument and you know it. Starlink is under SpaceX. Do you also think that is wrong then too? They are effectively doing the same kind of thing.
Kuiper is not under Blue Origin, and there are no whispers of Amazon and BO merging. You're the one being disingenuous in suggesting that companies have to be merged to buy services from - or cooperate with - each other.
The biggest problem with satellite internet was the costs involved, which SpaceX has pretty much solved.
Datacenters in space, on the other hand, are a terrible idea because of the laws of physics, which will not get "solved" anytime soon. But don't take it from me, listen to this guy with a PhD in space electronics who worked at NASA and Google:
Check the authors history. They are both anti AI and anti Elon. I think I feel a lot more confident staying optimistic and assuming that the SpaceX and xAI team have done their research about this. I know a lot of people are heavily biased in this matter due to politically not liking Elon or not liking AI, but I also think it's fair to say these companies have many very smart individuals working for them. If they have come to the conclusion that this is viable, then I have much more faith in what they are saying over one guys opinion who is biased against them and saying it's a bad idea.
You're also passing these judgements without knowing their full plan. Maybe we only know one part of the plan and maybe other details have not been announced. They may have a much bigger plan for this than just the specific information we have.
Having previously criticized someone doesn't make your technical analysis biased. It just means you noticed similar problems previously. Conversely, "I used to support him so I'm not biased" is given unearned credibility when really it just means you were late to noticing the obvious.
Technical analysis most definitely can be biased due to political leanings. This is why there is the whole idea that research can often be bought and paid for to get the results you desire. Because they are biased with money. Certain ideas or theories of how things could be done could very easily be overlooked or excluded by someone trying to dig for reasons to say something won't work.
What I am saying is that clearly SpaceX/xAI feel that this is a viable option based on many experts research/facts that are more knowledgeable than a single bloggers opinion. If I am thinking rationally why would I choose to believe a single random person over a group of experts banking A LOT of money that they have a solution that works?
I guess you'll have to wait and see what ideas they have to deal with this. If they can't manage the heat they aren't going to spend billions launching these things just for fun.
Although I'm sure SpaceX would be a non-trivial loss, the most important idea - their truly reusable rocket -- is proven to the point where other people are assuming they should do that to make rockets, it's like if Benz' company goes bankrupt in 1899. In that universe the Mercedes probably never happens but the automobile idea is already a done deal.
What do you mean? SpaceX didn't invent the reusable rocket, and my understanding is that Falcon 9 is still not significantly more economical than disposable rockets, and that the main reason it's attractive is that it's not Soyuz-2.
I found that surprising, so I looked on Wikipedia.
Soyuz-2 capacity to LEO: 8,600KG
Falcon 9 capacity to LEO: 22,800KG when expended, 17,500KG when not.
Soyuz-2 Cost to Launch: $35 Million
New Falcon 9 Cost to Launch: $70 Million
Used Falcon 9 Cost to Launch: $50 Million (cost to SpaceX: ~$25 Million)
Soyuz-2 cost per KG: $4000 (data from 2018)
New Falcon 9 cost per KG: $964 when expended, $1250 when not.
Use Falcon 9 coster per KG to Customer: $893 when expended, $690 when not
So realistically, Falcon 9 is roughly 20-30% the price per KG when new, and dropping to a minimum of 17.25% of the price when used.
Plus you get a larger diameter payload fairing and the ability to launch a payload up to 4X the size.
I'm pretty sure that even used as an expendable rocket, 1/4 the price per KG (if you need the capacity) is a pretty significant improvement. Now I understand why satellite ride-shares are so popular!
Space is basically half the cost of it's competitors on a per kg basis. And while previous experiments like the DC-X existed, SpaceX absolutely gets credit for the first operational reusable rocket stage.
And I say that as someone that despises Elon and the way he casts his companies as due to his personal technical genius.
It was a spaceplane and also a rocket. It literally had fixed rocket engines and was carried up by separating rocket stages. And yeah, it was expensive to operate, but it was built in the '80s and it truly was the first reusable rocket regularly flown, rather than being merely an experimental craft.
1. Elon is a genius, a real world Tony Stark.
2. How dare you! You're just jealous!
3. Ok, regardless, he's done more to advance EVe and space travel than anyone else alive.
4. Oh God, he's going to cripple US development of EVs and rockets, isn't he?
5. Eh, Mars was never happening in my lifetime anyway.
I think he's genuinely changed for the worse, quite a lot, in the last 10 years. Staring down failure seemed to keep the worst tendencies in check, being untouchable amplified them.
such a hilarious comment / mindset. he made the best selling car in the world 3 years running. neuralink is a great breakthrough. there are a string of accomplishments which individually would be the greatest thing many many people have ever done.
> he made the best selling car in the world 3 years running
Not only did Elon not found Tesla[0], but many employees have described the "babysitters" or "handlers" who are responsible for making him feel like his ideas have been implemented, so that his caprice and bluster don't interfere with the actual operation of the company.
To give him his due, he's a phenomenal manipulator of public opinion and image, and he certainly has invested a lot of his emerald-generated wealth into numerous successful ventures - but he himself is not a positive contributor to their success.
I mean, even if he isn't directly making a lot of the decisions in these companies that are doing well, it doesn't mean he doesn't play a big role in that still. He still had to pick a lot of these leaders, pay them well, keep them satisfied enough to stay there, and also give them the proper freedom to lead these companies. There are many people out there who could also manage to make these companies fail instead of grow.
I feel that a lot of people simply don't like Elon because of political reasons which are often also based on misinformed opinions. It also can't be denied that he is an intelligent person. You can hear it when he talks in interviews.
Now I think ultimately any ultra wealthy person is going to have some flaws that people can find and latch onto in order to hate someone.
Trying to make a point out of whether he did or didn't found Tesla kind if defeats the rest of your post. He paid over 90% of the first funding round and brought in key people like JB Straubel. When the company was basically an incorporation paper and no assets. Under most companies people would have argued for founder/co-founder status at that point. So yes he didn't "found" Tesla but for all intents and purposes he basically did.
I was shocked to learn recently how China is crushing it in renewables and electric cars. BYD sold 600,000 more electric cars than Tesla in 2025, becoming the world's largest EV brand. Tesla's sales have been declining since 2023, while BYD sales are rapidly growing, so the gap is likely to get even larger in 2026. This is an important trend, regardless of how one feels about Musk.
Hiding losses? From whom? He's the majority shareholder of both businesses. The combined company will go public and report on things like revenue, burn rate, etc. It's not financial engineering. It's a purchase.
Just say "rocket man bad" and save some keystrokes.
Legitimately, did you not immediately conclude it was for financial shenanigans? What did you think? I'm not trying to be shitty, but what else could there be?
Well if they plan to put a crap ton of new satellites in the air specifically for running xAI on it, I think there is a decent chance that it isn't purely financial shenanigans. Obviously the finances are probably a big part of such a decision, but companies also do these kinds of things all the time. I don't see why this is considered "shenanigans" or how the quote would relate to what is happening.
I'm gonna be honest, I really don't see how these opinion pieces are relevant. There are a lot of smart people working at these companies and I'm sure they have done a lot of research and work into determining that this is a viable thing to try. I am going to put more faith into that than somebodies opinion online.
It seems like a lot of people are very biased on this topic and want to see this fail because of who the company is. This author of this piece you linked appears to be both anti-AI and anti-Elon for example.
We also are unaware if there is some bigger strategy at play here and a bigger vision then what is currently being shared. I like to see companies try to innovate and take risks. I would like to try and be optimistic about things.
> I'm gonna be honest, I really don't see how these opinion pieces are relevant. There are a lot of smart people working at these companies and I'm sure they have done a lot of research and work into determining that this is a viable thing to try.
Like "robo"taxi, right? A lot of smart people have been working on this at same company for decade+
> I am going to put more faith into that than somebodies opinion online.
There are opinions and then there are things you can review that are factual and based on laws.
> Like "robo"taxi, right? A lot of smart people have been working on this at same company for decade+
I'm a bit confused what you're trying to imply here. They have launched RoboTaxi's and recently have been removing the human safety monitors in them. Are you trying to imply this didn't take a lot of work from a lot of intelligent people?
No, I am trying to not imply but say that it doesn't work which is why the company is now pivoting away to "humanoid robots" and is slowly starting to stop making cars.
You're talking about two different things. Robotaxi's are different from the self-driving style features of a personal Tesla vehicle. You specifically said RoboTaxi but now are referring to a pivot related to their Tesla vehicles.
I am just talking about (non-existent and will never materialize or exist) robotaxi. not about “full” “self driving” features of regular teslas (I own one)
They exist and are operational. I am confused by what you are talking about. How can you say "non-existent" and "never materialize" when they are quite literally in operation as a service right now.
Literally none. Space is the worst possible place to put something that overheats already on earth. There's probably some synergy in the other direction (AI piloting of satellites or whatever) but that's marginal at best.
I've sure they've considered that in the engineering. For example, the solar panels would shade it. The space station has a cooling system in it. Musk's Starlink satellites don't seem to be overheating.
that's an amazing read, lots of concrete and convincing challenges; but otoh, technology is evolving at such a fast pace, maybe it is possible for breakthroughs that we couldn't imagine now to become reality sooner than we would have anticipated?
There's another way to look at it, though. If the data center satellites can be built and launched cheap enough, you can still come out ahead on performance/cost. I.e. if the space data center has 1/10 the performance of a ground one, and they can be built and launched for less than 10% of the cost, then you've got a business. And there are costs that won't be incurred - no electric bill, no cost for land, no charge for maintenance.
* no electric bill: if you use solar panels to provide your own power, you also have no electric bill on Earth.
* no cost for land: land in sunny places where crops don't grow (for instance) is good for solar power and very cheap compared to building out a datacenter
* no charge for maintenance: sorry, I really don't get this one. Why don't the computers in space need any maintenance?
> maybe I am a fool, does space-based AI make no sense at all?
I think it does, for what it’s worth if we are to extend intelligence (as we know it) and potentially consciousness out there into the galaxy.
Because of distances and time, it is unlikely that humans will populate the galaxy with biological offspring (barring some technical breakthroughs that we have no line of sight on).
AI, on the other hand, could theoretically populate the galaxy and beyond, carrying the human intelligence and consciousness story into the future.
No. Imagine if your computer was in space instead of being under your desk. Would that solve anything?
Orbit is a very inconvenient environment. It's difficult to reach so maintenance is a nightmare, it's moving all the time, there's nowhere to sink waste heat into, you have a constrained power budget, you have a constrained weight budget. The only things you want to put in orbit are things that absolutely can't go anywhere else.
A lot of people who are a little bit ignorant think it's really easy to cool things in space because space is notoriously very cold.
Physics, it turns out, is slightly more complicated than this and it turns out vacuum is an incredibly good insulator and more (much more) than offsets the temperature differential in terms of how easy it is to cool something.
Thinking a bit, ORBITAL ai makes little to no sense, nowhere to dump your heat, your gpus are going to be slag or only operate part of the time. But what if he put them on the moon? the lag time is what ~1.2s? That seems like an amount of time that a current AI query can take and still seem reasonable.
Not that I think it's anything but him allowing some investors to cash out when spacex goes public. Hell didn't he just shift 2billion from tesla to xai?
At the end of the day he will never see whatever bullshit he's peddling in the media about this sale his drug habit is going to kill him before then.
without having watched the Big Short or having read the article, my first impression from the quote is "Megacorporations are failing dramatically, and the billionaires at their helm are freely doing financial gymnastics to pull the covers over the eyes of shareholders, while gaming the system to fully circumvent taxes and regulation -- the people with the power to do anything about it (legislators and regulators) watch idly (maybe profiting), the oligarchs make off like bandits despite copious failures, and the end consumer/taxpayer is either robbed or clueless this is going on, but most likely both, when there was a world where accountability could have been had and the common man was treated better."
the article headline immediately screams "financial gymnastics" to me so the rest followed from the quote.
I see what you're saying, but I also know that companies do these kinds of things all the time. It's very normal for companies to move around like this if it results in better financials. I don't see how that really makes this "financial gymnastics". Pretty much every company out there does some funny things with the numbers in order to reduce their tax burden. I wouldn't doubt this is the same kind of thing. If xAI plans to launch a ton of servers into the sky, it kind of makes sense for them to be apart of a company they also own that just so happens to launch satellites.
Starlink is also a company under SpaceX. Would you argue that is also financial gymnastics? Is it much different from what Starlink does? Instead of launching satellites to be a world wide ISP, they are launching them to be an AI provider.
I just don't see how this compares to the quote, otherwise it would apply to so many companies, including other ones already under SpaceX.
To me this just doesn't seem related and seems like a pretty big stretch likely biased by people who dislike AI and Elon.
This merger smells like a bubble. Servers in space? They don't make enough to cover costs here on earth. Americans will be forced to bail this mess up because we need Falcon 9, Starship one, etc.
The military (and/or government) should keep paying in advance for anything they need from SpaceX and make sure other unsecured creditors are not tooo significant.
When it all goes bankrupt, they can pay off the bonds for x¢ in the dollar and own SpaceX.
Perhaps if the gov could organize a little better, they'd make sure SpaceX owed lots of taxes and put themselves in front of the queue for ownership and screw other creditors (especially foreign).
The magic is not money or subsidies. Boeing got far more and they produced bupkis. Its organizational excellence. I don't know if that would survive. I guess if they can keep Falcon 9 stable then its still worth something but I imagine the star employees who grind themselves to dust getting this stuff to work do it for the mission and would depart if this occurred. Would Falcon 9 fall apart of that happens?
> Americans will be forced to bail this mess up because we need Falcon 9, Starship one, etc.
Or they could just go with the competition. If it came down to propping up something, I don’t see much difference between propping up ULA, Blue Origin or SpaceX. In the current environment who gets propped up probably depends on who scratches Donalds back.
This could just be a gamble they hope pays off. Or maybe there is some kind of other bigger picture strategy at play here that goes beyond just the AI from space idea. I try to stay optimistic about innovations in tech and I like to see companies willing to try new things instead of just staying stagnant.
For example, I think the car market had become pretty stagnant with traditional car makers, and most electric cars they attempted to make sucked. Tesla making good desirable electric cars really pushed EV's into becoming more popular and having a better charging network. I think it would have taken much longer for EV's to start growing in popularity if someone wasn't willing to take a risk.
Are they going to be too early to the market for this kind of tech? Maybe. Is it going to end up being a waste of money? Yeah it totally could be. But at the end of the day I do like to see some risks being taken like this and it sucks seeing constant negativity whenever companies try something new.
It is pretty clear. Socialize the losses of Musk investments by recovery via SpaceX contracts, supporting the US space program and the new Golden Dome program.
He sold FSD for 12 years, now is going to sell a Dyson Sphere for the next 30. This guy makes Ponzi look like a street hustler.
I think the argument is that it's messed up that a large debt swap from xAI kept Musk's margin on Twitter from being called by his investors, and now that debt is being absorbed by SpaceX.
This thread specifically excluded the big investors, but they too have nothing but loss popping the bubble: Musk has been talking up the value of their investment. If they criticize in public, they’re just costing themselves money — much safer to sell and walk away.
Well, no, the worry is that xAI's bondholders, who are also Twitter's bondholders, will be indemnified from any loss on those bonds at public expense because they are now also SpaceX bondholders and SpaceX is a national security interest of the US.
I think unsavory business practices actually affect approximately everyone, even those not directly connected to any one particular instance of unsavory business practices.
Whoa, I had to do a double-take on the Dorsey mention -- like, didn't he take the money and run while laughing at the folks that overpaid? But it seems he's retained a 2.4% ownership stake in Twitter/X, according to Wikipedia:
Still, don't make the mistake I did, which was to read the above comment to mean "he put more money in at the time of the buyout", since he was called an "investor in X".
Serious question .. are you long on cash or what? What investment class is not overvalued? Gold is but a store of value and doesn't really grow - look at how high it has gone.
I really don't know what you're trying to say. From your comment alone the conclusion I drew was that we should spend more on the industries that makes physical instead of financial products, such as SpaceX.
The sequence of events: Elon doing a leveraged buyout of X, then xAI funding, then debt transfers to X, then the xAI–X stock deal. Now the proposed SpaceX–xAI merger appears to have shifted X’s financial burden from Musk personally toward xAI investors and, potentially, future SpaceX shareholders.
This is speculative, of course, but yeah seems likely.
Who is SpaceX’s biggest customer? And which industry are we being told by any number of governments around the world is the most importantly thing ever and must be subsidized and forced on people?
It is estimated that Starlink is, accounting for 70% - 80% of revenue. Sources: [1] and [2]
NASA is SpaceX's biggest external customer for rocket launch services.
Although NASA is SpaceX’s largest external customer for traditional launch services, the company earns far more revenue from Starlink customers (millions of subscribers). So overall Starlink itself is SpaceX’s biggest revenue generator and de facto largest customer segment.
NASA contracts alone have exceeded $13 billion since 2015, with $1.1 billion expected for 2025.
The U.S. Space Force awarded $845 million for 2025 and $733 million for 2024.
Commercial satellite operators are estimated to contribute between $2.5 billion and $3 billion in 2025.
SpaceX saves its biggest customer money by being the superior, cheaper launch option. The alternative was ULA, which was an extraordinarily expensive monster.
Please highlight the problems you have with how it pertains to this context, how the biggest customer is harmed.
What do you care if its private owners are willing to absorb the mess that is xAI?
Why not put these AI/PV installations somewhere out in the ocean instead? A tiny fraction of the energy required to ship them there, you can actually physically get there to fix them, can use seawater for cooling, can use existing Starlink for connectivity, etc. Why/how is space more economical than international waters?
I can't tell how many layers of sarcasm are here, but I just want to highlight that aktshually cooling in space is quite difficult because there is no convection, so the only cooling option is radiative. Which gets a bit hard when the satellite gets blasted by the sun.
The ISS doesn't have problems staying warm, it has problems cooling off.
Everything I've heard from Musk in the past decade has been against my will and has made me dumber. (no I do not care to verify or know whether the above is true)
Edit: ah fuck ya got me "the next book in SpaceX and xAI's mission: scaling to make a sentient sun to understand the Universe" what the cultish bullshit is this. In a just world investors would be fleeing in droves from this cuckoo behavior (I know xAI & SpaceX are private)
A large telecommunications satellite operates at about 15kW. A Blackwell GPU consumes 1kW so you would be at 15 Blackwells per satellite. The cooling surface needs to scale linearly so there is little return to scale.
tbh you could just combine them with starlink sats. didn't they just apply for (and get?) a license for 1 million sats? Stick a single racks worth of gpu power on those and hey presto you've just got yourself the largest ai cluster in the world by far.
There's an audience for this. People have been taking these vapid statements and his trouble with communicating as evidence of genius for well over a decade now.
My question is how they'll update the hardware, if AI processors get outdated every 4-5 years? Or they will keep outdated satellites just for routing and Internet-related services?
The reason is probably that Tesla is falling behind on EVs, or at least feels like they've juiced all they can from them at the moment, but advanced robotics is still on the upswing and probably is far from reaching its full potential. They have enough money that moonshots like these probably seem irrelevant at their scale.
As for the space datacenter idea, I think this is just a case of extreme marketing that Musk's ventures are so accustomed to. Making huge promises to pump their stocks while the US government looks the other way. When time comes for them to deliver on their promises, they've already invented ten more outrageous ideas to make you forget about what they promised earlier. Hyperloop as a viable mode of transportation, tunnel networks for Teslas, SpaceX vehicles as a mode of transport, X as the new 'everything app', insane timelines for a Musk-led human mission to Mars. They've done it all.
Presumably the cooling problem gets hand waved away as a technical detail, and the real selling point is data centers that aren't subject to any regional governments laws.
Send up a spacecraft with back-to-back / equal area solar panels and radiators (have to reject heat backwards, can't reject it to your neighboring sphere elements!). Push your chip temp as much as possible (90C? 100C?). Find a favorable choice of vapor for a heat pump / Organic Rankine Cycle (possibly dual-loop) to boost the temp to 150C for the radiator. Cool the chip with vapor 20C below its running temp. 20-40% of the solar power goes to run the pumps, leaving 60-80% for the workload (a resistor with extra steps).
There are a lot of degrees of freedom to optimize something like this.
Remember MoviePass, and how they were losing gobs of money by letting people see unlimited movies for $20/month?
It was so obviously stupid that a bunch of people went, "well, this so clearly can't work that they must have a secret plan to make money, we'll invest on that promise", and then it turned out there was no secret plan, it was as stupid as it looked and it went bankrupt.
The "datacenters in space" thing is a similar play: it's so obviously dumb that a bunch of smart people have tricked themselves into thinking "wow, SpaceX must have actually figured a way it can work!"; SpaceX has not and it is in fact exactly as stupid as it looks.
But it won't end the same as MoviePass until Elon dies; he will keep moving things around, propping up failures with VC, IPO, federal/state (taxpayer) and profit making business money.
The pre-IPO timing of this narrative, combined with Musk’s history of Tesla’s stock pumping, leaves no room for doubt. But, that has worked for Tesla, I’m pretty confident it will work for SpaceX, which will IPO for $1T+.
My guess would be just a regular radiator and cooling system like a liquid pump. The only obstacle should be the vacuum. That said I don't have any hopes Elon has any understanding of any of it.
Cooling a datacenter in space isn't really any harder than cooling a starlink in space, the ratio of solar panels to radiating area will have to be about the same. There is nothing uniquely heat-producing about GPUs, ultimately almost all energy collected by a satellite's solar panels ends up as heat in the satellite.
IMO the big problem is the lack of maintainability.
> Cooling a datacenter in space isn't really any harder than cooling a starlink in space
A watt is a watt and cooling isn't any different just because some heat came from a GPU. But a GPU cluster will consume order of magnitudes more electricity, and will require a proportionally larger surface area to radiate heat compared to a starlink satellite.
Best estimate I can find is that a single starlink satellite uses ~5KW of power and has a radiator of a few square meters.
Power usage for 1000 B200's would be in the ballpark of 1000kW. That's around 1000 square meters of radiators.
Then the heat needs to be dispersed evenly across the radiators, which means a lot of heat pipes.
Cooling GPU's in space will be anything but easy and almost certainly won't be cost competitive with ground-based data centers.
Sure, but cooling a starlink in space is a lot more difficult than cooling a starlink on earth would be. And unlike starlink which absolutely must be in space in order to function, data centers work just fine on the ground.
You can just use the cheap solar panels that were gonna be launched into space (expensive) and not launch them into space (not expensive) and plug them into some batteries (still, cheaper than a rocket launch)
I think that it's not just about the ratio. To me the difference is that Starlink sattelites are fixed-scope, miniature satellites that perform a limited range of tasks. When you talk about GPUs, though, your goal is maximizing the amount of compute you send up. Which means you need to push as many of these GPUs up there as possible, to the extent where you'd need huge megastructures with solar panels and radiators that would probably start pushing the limits of what in-space construction can do. Sure, the ratio would be the same, but what about the scale?
And you also need it to make sense not just from a maintenance standpoint, but from a financial one. In what world would launching what's equivalent to huge facilities that work perfectly fine on the ground make sense? What's the point? If we had a space elevator and nearly free space deployment, then yeah maybe, but how does this plan square with our current reality?
Oh, and don't forget about getting some good shielding for all those precise, cutting-edge processors.
Assuming you can stay out of the way of other satellites I'd guess you think about density in a different way to building on Earth. From a brief look at the ISS thermal system it would seem the biggest challenge would be getting enough coolant and pumping equipment in orbit for a significant wattage of compute.
Because the latencies required for modern AI training are extremely restrictive. A light-nanosecond is famously a foot, and the critical distances have to be kept in that range.
And a single cluster today would already require more solar & cooling capacity than all starlink satellites combined.
Because that brings in the whole distributed computing mess. No matter how instantaneous the actual link is, you still have to deal with the problems of which satellites can see one another, how many simultaneous links can exist per satellite, the max throughput, the need for better error correction and all sorts of other things that will drastically slow the system down in the best case. Unlike something like Starlink, with GPUs you have to be ready that everyone may need to talk to everyone else at the same time while maintaining insane throughput. If you want to send GPUs up one by one, get ready to also equip each satellite with a fixed mass of everything required to transmit and receive so much data, redundant structural/power/compute mass, individual shielding and much more. All the wasted mass you have to launch with individual satellites makes the already nonsensical pricing even worse. It just makes no sense when you can build a warehouse on the ground, fill it with shoulder-to-shoulder servers that communicate in a simple, sane and well-known way and can be repaired on the spot. What's the point?
Isn't this already a major problem for AI clusters?
I vaguely recall an article a while ago about the impact of GPU reliability: a big problem with training is that the entire cluster basically operates in lock-step, with each node needing the data its neighbors calculated during the previous step to proceed. The unfortunate side-effect is that any failure stops the entire hundred-thousand-node cluster from proceeding - as the cluster grows even the tiniest failure rate is going to absolutely ruin your uptime. I think they managed to somehow solve this, but I have absolutely no idea how they managed to do it.
No they didn't. 200Gb/s is 25GB/s, so... They could run 1/36th of a single current-gen SXM5 socket. Not even any of the futuristic next-gen stuff. 25GB/s is less than the bandwidth of one X16 PCIe3 socket. And that's already assuming the best-case scenario, and in reality trying to sync up GPUs like that would likely have loads of other issues. But even just the sheer amount of inter-GPU bandwidth you need is quite extreme. And this isn't some point-to-point routing like Starlink trying to get data from A to B, this is maintaining a network of interconnected systems that need to communicate chaotically and with uneven demand.
According to Gemini, Earth datacenters cost $7m per MW at the low end (without compute) and solar panel power plants cost $0.5-1.5m per MW, giving $7.5-8.5m per MW overall.
Starlink V2 mini satellites are around 10kW and costs $1-1.5m to launch, for a cost of $100-150m per MW.
So if Gemini is right it seems a datacenter made of Starlinks costs 10-20x more and has a limited lifetime, i.e. it seems unprofitable right now.
In general it seems unlikely to be profitable until there is no more space for solar panels on Earth.
But it is always going to be significantly more expensive than a terrestial data center. Best-case scenario it'll be identical to a regular data center, plus the whole "launching it into space" part. There's no getting around the fuel required to get out of the gravity well. And realistically you'll also be spending an additional fortune on things like station keeping, shielding, cooling, and communication.
There are regularly comments on HN from people who say they pay for Musk's various products, and I am always downvoted into oblivion for suggesting that that is ethically problematic.
There's obviously quite a lot of autocratist illiberals in tech.
I am concerned, and haven’t seen anyone else point this out yet, that Musk will move Grok’s CSAM generation capabilities to space to be beyond the reach of terrestrial policing. Does this create some sort of legal loophole here so Musk can do this with impunity?
No. For one, things in space are under the jurisdiction of the country they were launched from. For another, it's people that do crimes, not satellites or LLMs, and the people involved in making CSAM are all on Earth.
AFAIK he can do whatever he wants in space, but CSAM is still illegal to view or even download in most (all?) countries of the world. So unless the degenerates also move out into space (which I'm sure they're eager to do), it wouldn't really ease the legal situation here on Earth.
Maybe if Elon launched himself and the dev team into orbit and didn't use any ground stations and just Starlink terminals he could start getting into legal loopholes.
But the "fraud" here is being done mostly to VC investors with deep pockets and lawyers, at least until he tries to take this entity public. And I can't imagine them just taking this lying down, but then again maybe they realize that offloading this steaming pile on public market investors is the best way out. But even then... SpaceX seemed like it was quite viable on its own, the investors there are the real losers here.
Anyone remember the quote by Russ Hanneman on SV [0] - "No Revenue, means you're potential pure play"
We know datacenters in space - sound plausible enough - yet not practical - hence they're potential pure play - also you can have massive solar in space - unlimited space -- etc -- all true -- but how economical / practical is it ?
yet we know on earth - to power the whole earth with solar - only a fraction of the land is needed. Hell it's even in the Tesla Master Plan v3 docs [1] - current limitation being storage & distribution
so all you - are now witnessing to the greatest scam ever pulled on earth.
The entire thing is a play. Musk should be a science fiction writer. He has that uncanny ability to create a statement that compresses 100+ years of industrial evolution into a few sentences.
>Factories on the Moon can take advantage of lunar resources to manufacture satellites and deploy them further into space.
I love how he goes from "the raw material is there" to "we will build high-tech supply chain to process them", just like that, magically.
Related: NASA chief suggests SpaceX may be booted from moon mission [1]. Blue Origin could snatch SpaceX's Starship lander contract. This looks increasingly a good idea.
Sean Duffy is no longer acting administrator of NASA. This proposal was apparently part of a bid to get the support of a coalition of old-space companies and new-space non-SpaceX companies. As part of that strategy he apparently leaked Isaacman's Project Athena document and was backgrounding that he was a SpaceX plant.
But, Isaacman is administrator now, and whatever you think about Isaacman and his relationship to SpaceX, I don't think there's much merit in thinking one of Duffy's half thought out plans is likely to be carried out.
Sadly this seems correct. When Trump was re-elected Elon Musk pushed for Jared Isaacman to be appointed as NASA administrator. When the pick went another way, it led to some real friction between Musk and Trump. Now, with Isaacman finally at the helm of NASA, it looks like Musk’s influence over the agency has come full circle.
>Is financial fraud consistent with our national prestige?
You're right, that may be all we have left to show for it if people can't come up with something better.
Whether it's Musk or anybody else who's a real example of outright fraud, in a top position where honesty and straightforward dealing mean more than anything.
The original Moon mission was masterminded by a literal card-carrying ex-member of the Nazi party (Wernher von Braun) and the American public back then didn't seem to mind.
All rocketry was, back then. You wanted ballistic telemetry? If you didn't know someone who worked on the V-2, you had to launch your own sounding rockets.
I think the parent's point stands. There's a lot more pragmatic concern with the damage SpaceX could do in 2026, versus the damage Nazis could do in the 1960s.
1. What in the circular funding? This feels more like a financing scheme founding it under X/Twitter and then spinning it over to SpaceX. I suspect some debt is disappearing or taxes aren't getting assessed because of this move.
2. The only thing harder than harnessing "a millionth of the sun's power" on Earth would be launching enough material into space to do the same thing. And that's not even a reason for SpaceX to own an AI company, at least not at this point. The current AI isn't going to help with the engineering to do that. Right now hiring 20-somethings fresh out of college is way cheaper and SpaceX has been very successful with that.
quick edit: dang, I even got point 1 backwards. xAI owns X/Twitter, and that means that SpaceX now owns X/Twitter as well as an AI company. Super suspicious that SpaceX could actually think that buying the social media part (a significant portion of xAI's value) would be worth it.
Seems like a way to put a lot of junk in space. If thats in earth orbit it will lead to a lot of junk falling from the sky in 10 years. If it all burns up that will be a lot of nasty shit in the atmosphere - millions of tons!
The really skeptical take here is that eventually all of Musk's companies merge, or at least the biggest ones, for juicing that market value to get that $1T payout. Looking at Tesla.
I thought xAi previously merged with twitter, so all of this is now rolled into SpaceX? Atleast the investors in xAi and the original financiers in twitter get a breather. SpaceX is the new bandaid for this hot mess. Let’s see if this ends up rotting SpaceX or if it gets healed.
In the past 40 years we’ve seen power per unit of compute decrease by over 40 million times. This says we need to put data centers in space because we can’t produce enough power on earth for AI. That won’t be the case as history has shown, but this is a great way to get AI money for your space ships if you’re going to IPO.
It's interesting as the space fans (who often dismiss AI) say this is a project to get SpaceX's space/Starlink profits for AI. But the AI people seem to think this is to get AI money for space projects.
Kind of a bad look - but I can't precisely say why. Maybe he thinks he can raise more capital this way than he could for each company separately? Especially raising more money for X might be quite hard - they seem to be quite a bit behind on the revenue side compared to OpenAI / Anthropic. With both companies merged he might just find enough retail investors willing to buy at sky high valuations.
Urgh... Elon is famous not for things he's done, but for saying things he is going to do. Do people still buy in to this? Elon Musk always promises things far in the future but doesn't make good on them. He hasn't succeeded in self driving cars. He is never going to mars. He is not solving the LA traffic problem with tunnels. His robots are the equivalent of the Metaverse. He's a phenomenal businessman, and understands that a story is part of that.
This makes me genuinely sad. SpaceX was the one thing of his that Elon has largely avoided screwing up. Imho, this is in large part due to Gwynne Shotwell. She seems to have the personality (not to mention, personal wealth) to kick Elon in the head when he tries to mess things up.
What’s happening now is nothing more than a transparent effort to couple the AI hype-wagon to SpaceX in order to drive the valuation higher in the minds of investors who still think that LLMs will completely transform society.
I’ll be thrilled if the rocket folks can avoid being distracted by this nonsense, but I’m not optimistic.
I’ve been following SpaceX since something like the 2nd Falcon 1 launch and this is the worst thing I’ve seen happen. Sad times.
What a clever trick to throw more money (governmental subsidies) into a sinking ship (xAI and "AI" in general). Perplexingly this maneuver will probably boost stock prices thus creating more monopoly money to burn resources with.
Doesn't the idea of Orbital Datacenters imply that the constraining resource right now is physical space, and not compute, electricity, etc?
Did we suddenly solve the electricity problem, or the compute problem?
As far as im aware there are still plenty of datacenters being planned and built right now.
Elon didn't want to get outshined after Sam Altman suggested that "maybe we build a big Dyson sphere around the solar system". When will people realize that these "geniuses" are only good at making money, and any benefit to society is coincidental.
This is very clearly a move to sell SpaceX IPO at overinflated prices by promising a vision of a future which is perpetually 5 years away. SpaceX is going to be the next meme stock.
They must have linked the wrong press release /s. I would have expected a press release about SpaceX acquiring xAI to talk about why they did that. Or at least mention xAI beyond the first paragraph. This is just Elon talking about space data centers
I'm not sure if this is meant to be sarcasm but is there really a need in the car market for on-demand CSAM? What actual use does Grok or any LLM have in a car?
With that number of more satellites in orbit, launching a manned rocket into space is likely to be too risky due the amount of more debris and satellites.
And will it be net energy positive solution ?
One thing to keep in mind. xAI and SpaceX both have contracts with the DoD. So it makes sense he moved it there rather than Tesla. Not sure I buy the needing AI for doing more in space or if this is to save sinking ship, but if one of his two big companies needed to buy it to keep it afloat it makes sense it was SpaceX and not Tesla.
I'm wondering if SpaceX's going public will be delayed. If not we'll see the first test of the public's appetite for what the AI companies' balance sheets look like
Whenever computer chips go into space, they have to be hardened against radiation, because there is no atmosphere to protect them. Otherwise you get random bit flips.
This process takes a while, which is partly why all the computers in space seem out of date. Because they are.
No one is going to want to use chips that are a many years out of date or subject to random bit flips.
(Although now it got me thinking, do random bit flips matter when training a trillion parameter model?)
That's not exactly how LLM temperature works. :). Also that's on inference, not training. Presumably these would be used for training, the latency would be too high for inference.
Latency would be fine for inference, this is low earth orbit, that is about 25ms optimistically. Well within what we expect from our current crop of non local LLMs.
Speculation - Starship is a dead end all of this is a play to inflate IPO prices for SpaceX during the AI bubble. Then cash out and start a new private space company?
> Starship will deliver millions of tons to orbit and beyond per year
Excuse my naive physics, but is there a point at which if you take enough mass off of earth and launch it into space, it would have a measurable effect on earth's orbit? (Or if the mass is still tethered to earth via gravity, is there no net effect?)
Earth weighs about 5972200000000000000000000 kg. They are claiming to plan to launch ~5000000000 kg / year. That's 8x10^-14 %. You'd need some pretty accurate instruments to tell the difference.
"Launching a constellation of a million satellites that operate as orbital data centers is a first step towards becoming a Kardashev II-level civilization"
So, basically give ourselves Kessler syndrome. Or is Elon trying to monopolize orbit entirely?
Let's call it for what it is, a payday for Elon. Paper billionaires have figured out they cannot cash out with out tanking their paper, so now you have these circular deals to extract as much as possible. If we had a functioning government they would step in and put an immediate stop to this on national security grounds.
Didn’t Elon say that orbital solar collection was a stupid idea due to energy loss in transmission? Using AI as an almost proof-of-work shows that it may potentially be more complex problem than previously thought. If we threw Bitcoin miners up to those satellites you could literally beam money down.
Musk is moving value out of public hands and into his own. He overpaid $44B for Twitter, then rebranded it as an AI asset by folding X into xAI. He pushed Tesla to invest $2B of shareholder money into xAI despite shareholders voting no. Five days later, SpaceX acquired xAI, effectively turning Tesla’s cash into equity in a private company Musk owns far more of. Musk controlled every step, there was no real arm’s-length process, and he almost certainly knew the outcome in advance. Musk and his private investors get control, inflated valuations, and IPO upside. Tesla shareholders supply the cash, take the risk, and lose leverage.
If Musk and SpaceX are serious about putting 1 million datacenter satellites into space, then they are not serious about Mars.
You cannot simultaneously build and launch 10’s of thousands of Starships to deliver 1 million tons of equipment and supplies bound to Mars while also committing to launching 10’s of thousands of Starships to orbit full of satellites.
They would need to quadruple their launch rate, and half of those launches would be Starships bound for Mars, the vast majority of which would never return.
How many Falcon9’s have ever been built? It is incredible to say you can build that many rockets and use up that much fuel on any reasonable time scale. You might as well say the Tesla Roadster version 3 will be a Single Stage to Orbit rocket car.
well, Musk has been overpromising and under delivering for a decade (or more?), so it seems pretty clear this too is shithousery, albeit possibly ambitious.
I suppose one of the ADR’s read something like “…who cares about bitflips, man. Isn’t AI all about probability?”
Knowing the insane level of hardening that goes into putting microcontrollers into space, how to the expect to use some 3nm process chip to stand a chance?
A trend at the moment is to just hope for the best in cubesats and other small satellites in LEO. If you’re below the radiation belt it’s apparently tenable. I worked somewhere designing satellite hardware for LEO and we simply opted to use consumer ARM hardware with a special OS with core level redundancy / consensus to manage bit flips. Obviously some problems will present for AI there… but there are arguably bigger problems with AI data centres like the fact that they offer almost no benefit with respects to the costs of putting and maintaining stuff in space!
I have so many conflicting thoughts that I cannot properly articulate yet. I can say though, this is not going to end well for most, it is clumsily premeditated and starting to feel like dude is just trying to be a Neal Stephenson character.
Just checking (genuine question) there wouldn't be a sneaky way to weaponize a million satellites in orbit around the Earth, would there? I can't imagine it wouldn't have ever been looked into.
I don't see the demand for space being there, OSS is driving costs down and there are still plenty of hardware and algorithmic optimizations we haven't deployed yet.
I will not be left holding this bag. This is such financial engineering nonsense, and if we had any sort of regulatory controls this would never be allowed to happen - especially BECAUSE of national security reasons.
Do this math include the cost and weight of the radiators? Because it obviously can't work without big radiators, and I don't see them mentioned in the math?
This makes a lot of sense. The commercial launch business is not large enough to support all possible Falcon launches, so Starlink was created to take advantage of the low launch cost and vertical integration and is now a major profit center for SpaceX.
Starship launches are only going to make sense every 779.94 days (the approx 2 year Mars-Earth proximity). The rest of the time, the launches could similarly be used to deploy orbiting data centers for XAi/Grok etc. Brilliant move.
SpaceX launches about 60% of National Security Space Launch payloads. The only other active launcher cleared for these missions is ULA's Vulcan Centaur, which has launched 3 times total, ~once every 8 months.
I'm not really interested in getting in a political mudslinging contest; you asked what's to stop a future president from doing so and I gave a practical answer.
I really would like to see a cost and cooling breakdown. I just can't see how you can do radiative cooling on the scales required, not to mention hardening.
I thought this was a troll by Elon, now I'm leaning towards not. I don't see how whatever you build being dramatically faster and cheaper to do on land, even 100% grid independent with solar and battery. Even if the launch cost was just fuel, everything else that goes into putting data centers in space dwarfs the cost of 4x solar plus battery.
I cannot really tell satire apart from genuine opinions anymore.
(But I do hope it was satire, if not, cooling satelites was/is a big issue and they only have very modest heat creation. A data center would be in a quite different ballpark)
Pre-IPO price padding. xAI is going nowhere but at least for now it has some value. Move it under SpaceX, bump up SpaceX’s valuation and therefore it’s opening IPO price. Then kill xAI and write it off.
This is terrible for Space-X. They're doing a great job. Musk has left running it to Gwynne Shotwell, who really is a rocket scientist. Now Space-X has a AI business unit they don't need, a new money drain, and more attention from Musk.
Should have merged xAI into Twitter. A failure there would not be a major setback.
I know that per HN's guidelines we're supposed to be "kind and curious", and "reply to the argument instead of calling names". But with some texts, engaging with individual arguments loses sight of the more important bigger picture. So while unkind, the most "thoughtful and substantive" thing I think can be said about this text is:
Friendly reminder for anyone that forgot - xAI acquired Twitter, so now Space-X is the proud owner of a dying social media platform that they overpaid for.
Any claims that this is about putting compute in space is just a non-sense distraction. This was absolutely about bailing Elon out of his impulsive, drug-fueled Twitter purchase.
The only question now is: when they try to go public, will they be punished for wasting so much money or not? My guess is: not.
People confuse being able to think big with being allowed to think big. With this much money loaned to him, protectionism and too many PR stunts, I don't think he delivered big enough, and he won't be able to at some point. This is madness.
I think there is one more possible framing for this.
Recently xAI has been in the news for Groq's revenge-porn-like "undress them" feature, which seems pretty legally questionable.
Musk has also been in the news for his own Epstein-related activities.
If he can move Groq and X into space, well, there's not very many age-of-consent or revenge-porn laws in space as far as I know, so maybe he'll be able to do some sort of legal leverage where the space data-center can produce otherwise legally questionable AI responses with impunity.
Sorry but who buys this bullcrap? Sure you can take a tesla and drive it in your bedroom to use its AC, because you make teslas, but you could also use just a window attached AC?
So they use a valid and valuable company to hide a giant dumpster fire company. To add to that, their best argument is "AI in space", which has some real "solar roadways" energy to it. I honestly don't know how any SpaceX shareholder could approve this.
Musk’s schtick lately has always been to find a challenging problem, point everyone to it and say “ I will solve it by September “, have the stock shoot up and make money. His first dis that with self driving, then twitter, then xAi, and now robots and data center in space. These last two will last him a decade as these are both challenging problems to solve over night.
We just had an X8.1 CME event, I just want to point out that at any moment we could have an x40 (we had a carrington event already in 1859) or higher event and all those sats at low earth orbit would be fried and start hitting each other, if SpaceX keeps launching more it becomes incredible probable that we might hit the Kessler Syndrome, and we would legit lose access to Space for a WHILE, including all of what satellites entail.
Are we ready for that as a modern society or are we going to start enacting regulation against it? I'm sorry but people wanting internet everywhere does not justify we going back to the dark ages for a decade or more.
US is dialing back even on global warming. There is no chance current government would risk space superiority for some kessler-shmessler that nobody has ever seen.
Pretty terrible for SpaceX. Of course they paid a crazy inflated price for xAI in an attempt to cash in on the IPO. This just devalues SpaceX and exposes the investors to all the AI bubble risk.
> SpaceX has acquired xAI to form the most ambitious, vertically-integrated innovation engine on (and off) Earth, with AI, rockets, space-based internet, direct-to-mobile device communications and the world’s foremost real-time information and free speech platform. This marks not just the next chapter, but the next book in SpaceX and xAI's mission: scaling to make a sentient sun to understand the Universe and extend the light of consciousness to the stars!
Tesla has probably the most valuable shareholders on Earth. Over years of empty promises and meme status, the stock has pretty much purged all the level heads. So it's mostly deluded Elon sycophants giving placing their tithe on the alter of his sci-fi fantasy smoke and mirrors game.
In reality he will be dumping the debt of twitter and xAI (and maybe spacex?) on Tesla shareholders, and buoying that with the added layer of hyper that spaceX brings.
Now he just needs to work in crypto satellites down to users via all the new phones supporting satellite link to SpaceX. I kinda expected that one first. Distributed payment network outside of government control/oversight seems like something he would be in to.
Reminder that space only allows for radiative cooling (since there is no air to absorb heat) so data centers in space are going to have massive cooling panels.
Hyperloop was never a company project, neuralink was a separate company, tesla is rolling out driverless robotaxis and fsd is amazing, robots inevitably are going to do the majority of work - there’s no real doubt about it is there?
> robots inevitably are going to do the majority of work - there’s no real doubt about it is there?
There's a lot of doubt that the AI and compute to enable that would happen on commercially relevant timescales.
Consider: "do the majority of work" is a strict superset of "get into car and drive it". The power envelope available for an android is much smaller than a car, and the recently observed rate of improvements for compute hardware efficiency says this will take 16-18 years to bridge that gap; that plus algorithmic efficiency improvements still requires a decade between "car that can drive itself" and "android that can drive a car". (For any given standard of driving).
And that's a decade gap even if it only had to do drive a car and no other labour.
You can't get around this (for an economy-wide significant number of androids) by moving the compute to a box plugged into the mains, for the same reason everyone's current getting upset about the effect of data centres on their electricity bills.
And note that I'm talking about a gap between them, not a time from today. Tesla's car-driving AI still has safety drivers/the owner in the driving seat, it is not a level 4 system. For all that there are anecdotes about certain routes and locations where it works well, there's a lot of others where it fails.
That said: Remote control units without much AI are still economically useful, e.g. a factory in Texas is staffed entirely by robots operated over a Starlink connection by a much cheaper team in Nairobi.
I appreciate you engaging, but I'm not sure power how would be the limiting factor. Assuming an average of 1kW of compute needed per robot (for reference, Tesla's AI4 is ~200W, rumors say 800W for AI5, nvidia B200 is ~1kW), that's nothing compared to the amount of energy we use for locomotion (a car eats like 20kW at 60mph).
Reminder that SpaceX has received an estimated $38 billion in government funding over the years, and all of its returns are going to a small set of private investors.
Socialized losses, privatized profits. As is the American way.
Man the real story in a few decades is going to be whether SpaceX was onto such a kille business that it survived being used asma fiscal dumping ground for the losses being incurred by mismanagement everywhere else.
xAI to cover X investors, SpaceX to cover xAI, us American public investors to cover SpaceX since it cannot go under for strategic reasons. Ultimate grift.
Major grift vibes, inventing half-baked reasoning to justify massive valuations.
If money wasnt so deeply entwined with politics at this point, this is the sort of news that would launch fraud investigations.
The way I read it, X AI is not really profitable and Elon's creditors/coinvestors asked for something tangible for their money, a.k.a shares in SpaceX, his only business that still has some solid foundation. The rest is emois.
As a fan of space this makes me sad. AI Datacenters are complete nonsense. And binding all of this stuff SpaceX is idiotic. SpaceX could do so many great things for space and now its all messed up in the AI race.
When Elon Musk talks about benefiting humanity, remember that the one time he had unregulated power (DOGE) he used it mostly to cut benefits for poor people, and to push ideological agendas. His only agenda is self-aggrandizement, and this announcement is cover for passing the hot potato of Twitter debt around.
Whenever Elon, Anthropic or whoever mentions humanity, they don’t mean every 8 billion of us on Earth. They mean themselves, their mates and whoever they deem worthy of the posterity they believe only they have the merit to design.
I thought this wasn't viable due to cooling requirements - how do you cool massive amounts of compute when the only option is to radiate it into space - nothing to convect it with?
Also, the incredible amount of grift here with the left hand paying the right is scarcely believable. Same story as Tesla buying Solarcity. Board of directors should be ashamed IMO.
Yes. It is very cold up there but there is also no matter, or very little matter. So head conduction and convection don't work, it's all radiation. When we are learning to solve heat transfer problems in engineering school we are generally taught to neglect radiation, because it's effect on cooling the system is typically second or third order when compared to the to "big C's"
Not my experience at a hyperscaler, at least a while back. It definitely made financial sense to swap a small part to get a ~50-100k$ server's capacity back online.
The overall sentiment of the discussion is *overwhelmingly skeptical and critical*. While a small minority of users defend Elon Musk’s track record of defying critics (citing reusable rockets and EVs), the vast majority view the "AI datacenters in space" proposal as scientifically unsound and economically nonsensical. Most commenters interpret the merger as a form of "financial engineering" designed to bail out underperforming assets (Twitter/X and xAI) using SpaceX's valuation ahead of a potential IPO.
The most robust debate focused on the physics of operating high-density compute in space.
* *The Cooling Problem:
* Numerous engineers pointed out that space is a vacuum and therefore an excellent insulator. While solar panels generate power, getting rid of waste heat from GPUs requires radiative cooling, which is inefficient compared to terrestrial convection (air/water). Users estimated that the radiator surface area required would be massive and structurally prohibitive.
* *Radiation & Durability:* Commenters noted that cosmic rays cause bit-flips and degrade electronics. Terrestrial hardware (like standard GPUs) would not survive long without heavy, expensive shielding, or would require "space-grade" hardening that lags generations behind in performance.
* *Maintenance:* On Earth, failed components are swapped; in space, a failed GPU or drive effectively bricks the unit or turns it into space junk.
* *Latency:* While Low Earth Orbit (LEO) offers better latency than Geostationary orbit, users questioned the utility of high-latency inference compared to fiber-connected terrestrial centers, particularly for complex AI tasks.
A significant portion of the thread analyzed the merger as a financial maneuver rather than a technological necessity.
* *The "Bailout" Theory:* Users widely believe this deal is designed to offload the heavy debt and losses from the Twitter (X) acquisition and the high burn rate of xAI onto SpaceX, which is viewed as Musk's most solvent and valuable company.
* *SolarCity Parallel:* Many compared this to Tesla’s acquisition of SolarCity—a move previously criticized as a bailout of a failing Musk-owned company by a successful one.
* *IPO Preparation:* Speculation suggests this is a play to juice the valuation of a SpaceX IPO by attaching "AI hype" to it, allowing early investors in the struggling X/xAI entities to cash out or convert their equity into SpaceX stock.
* *Conflict of Interest:* Commenters questioned the governance of private companies where one individual (Musk) controls the board and directs mergers that may benefit him personally at the expense of specific shareholder groups (e.g., SpaceX employees or investors).
### *Category 3: Economic Viability*
Users attacked the business logic of launching datacenters into orbit.
* *Cost of Launch vs. Land:* Even with the cost reductions promised by Starship, users argued that land and grid connections on Earth are orders of magnitude cheaper than rocketry.
* *Solar Efficiency:* While solar is more efficient in space (no night/clouds in specific orbits), users argued it is still cheaper to simply build more solar panels on Earth and use batteries than to launch infrastructure into orbit.
* *The "Million Ton" Claim:* Users crunched the numbers on Musk’s claim of launching "a million tons" of satellites, noting it would require an unrealistic flight cadence (e.g., launching massive rockets every few hours continuously).
### *Category 4: Musk's Reputation & Rhetoric*
The thread discussed Elon Musk’s history of promises versus delivery.
* *Skepticism:* Users cited a long list of missed timelines and unfulfilled promises (Hyperloop, Full Self-Driving by 2017, Robotaxis, Mars landings) as reasons to doubt the "space datacenter" timeline of 2–3 years.
* *Mockery of Language:* There was specific ridicule regarding the press release language, particularly the phrase "scaling to make a sentient sun," which many found to be "drug-induced nonsense" or "cultish."
* *The Defense:* A minority of commenters argued that betting against Musk has historically been a bad idea, citing the success of Falcon 9 and Starlink as proof that he can solve "impossible" engineering problems.
* *Jurisdiction Shopping:* Some speculated that moving AI to space might be an attempt to bypass terrestrial regulations regarding copyright, safety, or content generation (specifically referencing Grok’s lack of guardrails regarding CSAM/deepfakes).
* *National Security:* Concerns were raised that SpaceX is a critical US defense contractor, and merging it with a "chaotic" social media company and an AI firm introduces unnecessary risk and leverage over the US government.
* *Orbital Debris (Kessler Syndrome):* Users worried that launching millions of tons of disposable datacenter satellites would clutter low earth orbit, increasing collision risks and potentially locking humanity out of space travel.
### *Category 6: The "Why" (Strategic Speculation)*
* *Energy Arbitrage:* A few users attempted to steelman the argument, suggesting that if Earth's energy grid becomes the primary bottleneck for AI, space offers the only unconstrained solar power source, despite the cooling difficulties.
* *Vertical Integration:* Some noted this creates a conglomerate similar to Samsung or aggressive Japanese keiretsu, where the goal is total vertical integration of energy, transport, communication, and intelligence.
Elon Musk is a genius, but he’s a financing genius. Look at the long history he has of false promises supporting financing deals between his companies and you’ll see this for what it is, a cash injection and a lie to justify it. He did the same thing with a fake solar roof demo when Tesla bought the almost bankrupt Solar City. He also shifted resources from Tesla and SpaceX to support X in the early days. Even founding xAI outside of Tesla, when so much of its valuation was built on its AI capabilities, was questionable.
When does the market realize this is all just a shell game and the emperor really has no clothes?
We saw this on a much smalelr scale a decade ago when one of Elon's companies (Tesla) acquired a second one of Elon's companies (SolarCity) because it was broke and owed a ton of money to a third one of Elon's companies (SpaceX).
Elon was forced to go through with his impulsive Twitter acquisition by a Delaware court, an acquisition that was not only secured by a bunch of Tesla stock but also a bunch of Qatari and Saudi royal money. He then mismanaged Twitter so badly Fidelity wrote down its value by at least 80% [1].
So what did Elon do? Raised even more questionable foreign money into xAI, diverted GPUs intended for another of his companies (Tesla) into Twitter and then "merged" Twitter into xAI, effectively using other people's money to bail him out from an inevitable margin call on his Tesla stock.
Interestingly, Twitter was reportedly valued at $33 billion in this deal [2], significantly more than the less than $10 billion Fidelity valued Twitter at. Weird, huh? With a competent government, this would be securities fraud that would have you spend the rest of your life in jail. And even with all that, $11 billion was lost on the deal.
So here we are and it's time for the shell game to be played again. Now it's SpaceX's turn to bail out the xAI investors.
And what is the argument for all this? AI data centers in space. Words cannot describe how little sense this makes. Launch costs (even if the Starship launch costs get to their rosy projections), cooling in space, cosmic rays (and the resulting errors) and maintenance. Servers constantly need parts replaced. You can just deorbit the satellite instead but that seems like an expensive way of dealing with a bad SSD or RAM chip.
Making a "sentient sun" is the most bald-faced asinine, drug-induced nonsense that should be the complete destruction of all credibility to anyone who said it or typed it or works for anything here.
Financial engineering. Twitter under Elon became a dumpster fire of porn and hate and big banks were holding 13B in bonds that wouldn’t be worth the paper they were printed on for the company alone so he just links it with his only company that actually is doing something worthwhile…
Not sure how X which “merged” wit X (formerly Twitter) and SpaceX really matter or synergize but here we are. It’s all about the money being protected. And this Ketamine using wierdo is gonna be the worlds first trillionaire. Yay all of us.
> By directly harnessing near-constant solar power with little operating or maintenance costs, these satellites will transform our ability to scale compute. It’s always sunny in space! Launching a constellation of a million satellites that operate as orbital data centers is a first step towards becoming a Kardashev II-level civilization, one that can harness the Sun’s full power, while supporting AI-driven applications for billions of people today and ensuring humanity’s multi-planetary future.
Apparently optimus robots don't work and he needs to start his final grift, space datacenters, while his datacenters on earth are powered by gas turbines.
Most likely he's just trying to bury his epstein involvement where was exposed lying by his own daughter.
Can someone convince me that this is not a) pure horseshit b) a plan for Elon to sneak enough mass into orbit to hold the Earth hostage?
If you can bring millions of tons of anything into orbit around Earth you can destroy civilization, or just France.
> This marks not just the next chapter, but the next book in SpaceX and xAI's mission: scaling to make a sentient sun to understand the Universe and extend the light of consciousness to the stars!
Slightly less high Musk aide: "But what is the synergy, where's the moat and how could that be done in practice and most importantly is there any limiting factor on Earth before we have to bring AI into.."
Musk : "SPACE!!!!!!"
It is incredible to think that the extremes of the stock market are actually pretty similar, pink sheets/cryptos and these mega companies are actually the same. News fueled pumps and dumps to win the cycle of hype of the week
Elon put everything he had into spacex and tesla. Against all odds, both worked out. He's now betting tesla on optimus. People are willing to believe him given his extremely successful track record of beating the odds. Is it guaranteed? No. If it was guaranteed it wouldnt be an investment. You take on risk for the chance at a reward. Calling it hopium is ridiculous--if there's anyone on the planet that has a track record of pulling off the near impossible its Elon.
There's an epic conflict of interest here with Musk owning most of both companies. And they're in entirely separate fields, there is no plausible synergy here to be gained.
How can you have a conflict of interest if they're entirely separate fields? They have different interests, so where's the conflict?
You don't need synergies to justify a merger. They're often used as justification as in paying well above market price. But it has nothing to do with actual justification. You can just have a holding company of businesses
Yeah but who can be hurt by this, these are both private companies? So whose interest is his "conflicting" with? I'm sure the shareholders will raise it with him and/or bring a lawsuit if they aren't happy (they probably are happy).
Yeah, I'm not buying that. I don't see how that could be any cheaper than regular datacenters. It might just be technically feasible, but launching stuff into space will always be more expensive than not launching stuff into space. And all those pesky technical issues like cooling might be solvable, but I doubt they're that cheap to solve.
No he is not. It makes no sense from a physics standpoint or an economic standpoint. And even if they were, it wouldn’t require whatever this acquisition is.
Musk has a history of having one of his more successful companies buying one of his less successful companies. xAI bought X, and Tesla bought SolarCity
Musk is notorious for shuffling assets across his companies to make some financials look better. For example, shuffling Twitter servers (and then all of Twitter) under xAI.
my partner got shingles a couple years ago, it was a very painful experience!
(to be crystal clear, I am making a joke equating the failed SolarCity/Tesla solar shingles to the (generally considered very painful) Herpes Zoster manifestation also called "shingles")
What's funny? Do you think the investors are against this? The investor's aren't idiots. I imagine the typical investor in Elon Musk's companies would approve of this sort of thing. So what's the problem? Besides, its a private company with Musk as majority shareholder in both. That's the beauty of private companies, you can just do things.
I wish more companies were private and ambitious. I'm tired of companies like Apple making marginal spec bumps to their phones and milking the same products for decades
More than 70% of voting shares supported the package, very close to the level of support in the original 2018 vote. This excludes Musks share.
And consider that this is retroactive, meaning it's backpay. They're literally voting to give the guy $50b for work performed. He has a lot of confidence from his investors. And if there were issues, there would be lawsuits. Ironically the only lawsuits that get brought up, like the one about the pay package, are basically trolls, from a guy that had 9 shares.
Besides the parent is the one making a claim that something not above board is going on so burden of proof is on him.
Finally, it's a private company where Musk is the majority shareholder. He's moving money from one pocket into another, and any moves will be reflected in his attempt to raise money with the IPO coming this year.
I can't imagine the world view you would have to hold to think that people who manage to command tens of billions of dollars to invest are idiots, just tripped over the money and just go off vibes.
The investors want to cash out, Musk needs lots of money to plow into his latest toy that so far only excels in ridiculing him and sexual harassment/CSAM, so they make a deal to take in xAI and go public. Win win.
It's widely reported that Musk is a majority shareholder of xAI and the controlling shareholder of SpaceX (close to 80% of voting shares). Not surprising that he would be looking to consolidate ownership under one entity especially if he perceives significant synergies (i.e., data centres in space).
Shocked to see SpaceX buy the datacenter in space meme. Where does the power come from? Where does the heat go? Why add (high) launch costs to your buildout capex? Why add radiation as another risk factor to your already-unreliable GPUs? Am I missing something fundamental here...?
Money! Also power source is just solar - not too difficult. I don't think radiation would be too much of an issue either since they're in low earth orbit. Heat is probably the biggest problem. Or manufacturing & launch costs. Pretty silly idea anyway.
Aside from Elon Musk, there are a few other people with a lot of capital aiming to do the same thing. That means, either they are all wrong (possible) or this problem has been solved somehow and the solution itself is not public.
Google and Amazon are doing the same thing. Maybe it is a moonshot (pun intended), but Musk is hardly alone in the push.
Not to mention the huge issues of cosmic rays. Sure, if the lifespan of the satellite is expected to be low, then maybe tolerable. But even then, how would this be financially viable?
I think it's far more likely that he wants to combine his businesses to roll his really expensive, debt-ridden companies into one entity with the company that actually reliably makes money.
Indeed. But it's also a hilariously Musky idea! Some moderate technical competence paired with sociopathy and an ego orders of magnitude too big, and voila, you get Cybertrucks, Hyperloops, Neuralinks, Teslabots, datacenters in space, and all the other garbage the man spews.
I cannot wait for him to one day be hit in the face by reality.
I have never understood how Data centers in space ever make economic sense, the payload, latency and many other issues make it difficult at least for the immediate needs
given the max bandwidth of a starlink sat is in the 100Gb on a good day range why would you want to limit a DC to less bandwidth than a single cheap fibre?
Also in LEO you're going to have reentry become more of an issue (starlink burning up in the atmosphere isn't some free garbage removal it will have a measurable impact on the chemical make up (assuming it even burns up and doesn't just squash more farm buildings), Power supply more of an issue and still have huge problems with heat and radiation.
wow this is one of the most incredibly fraudulent things that ever happened in American capitalism and I'm not ignoring Enron or the mortgage meltdown. I'm speechless the US has given up on any semblance of law and order in matter of financial markets and this stuff can happen without people going to jail.
This means that Grok, Elon's politically labotomized involuntary pornography generator, and X (formerly Twitter), Elon's Nazi-adjacent propaganda machine, are now completely intertwined with SpaceX, a too-big-to-fail government contractor that currently serves as America's only reliable option for manned orbital spaceflight.
Anyone who doesn't see how broken this situation is isn't paying attention. This is how people like Elon, who want to seize as much power from the government as they can, ensure that the means for seizing that power are untouchable.
Anyone who has ever used Grok or X lately knows that both of these products are heavily manipulated to align with the political, social, and economic views of Elon Musk, who is increasingly boosting "white power" language and full-throatedly backing America's most nationalistic and authoritarian president to date.
This is just another consolidation of power, and it's deeply worrisome. Any integrity one may have hoped remained at SpaceX just vanished when they aligned their mission with that of these deeply problematic digital services.
And this is not even scratching the surface of what looks like a deliberate attempt to create Kessler syndrome by launching millions of cheap short-term satellites into orbit, or the rationality of putting datacenters into orbit in the first place...
People pooh-poohing space datacenters will obviously think this is a bad move. But Elon clearly believes space datacenters will work. Given that, and the fact that SpaceX will IPO this year, this acquisition was inevitable.
SpaceX and xAI would not be able to freely collaborate on space datacenters after the IPO because it would be self-dealing. SpaceX likes to be vertically integrated, so they wouldn't want to just be a contractor for OpenAI's or Anthropic's infrastructure. Merging before the IPO is the only way that SpaceX could remain vertically integrated as they build space datacenters.
Most everyone here is just angry Elon opened the door for freedom of speech to exist online and started the process of uncovering the left's extreme corruption. They feel betrayed by him so he's enemy #1. Literally could cure cancer and he'd still be branded as evil by the left. Ironic as he's arguably done more for leftist causes (renewables, ev's, etc) than anyone else on the planet.
These are private companies, with private investors. If Elon really screwed all investors over and over again he wouldnt be able to raise money repeatedly. But he has no problem raising money. Because the claims that he's doing shady things or screwing investors is entirely unfounded. Same goes with the claims that he's an idiot or not that smart. If thats really the case, show me what you've accomplished with your genius.
It takes intelligence to recognize intelligence. An idiot cant recognize a genuis.
Elon's purchase of Twitter (now X) opened the door for conservative view points to be heard online. Prior to that, all other major platforms were censoring conservative views (with wide spread company leadership claiming due to pressure from Biden administration).
I regularly use X. I get blasted with political posts from both sides constantly on it.
> it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power
We currently make around 1 TW of photovoltaic cells per year, globally. The proposal here is to launch that much to space every 9 hours, complete with attached computers, continuously, from the moon.
edit: Also, this would capture a very trivial percentage of the Sun's power. A few trillionths per year.
We also shouldn't overlook the fact that the proposal entirely glosses over the implication of the alternative benefits we might realize if humanity achieved the incredible engineering and technical capacity necessary to make this version of space AI happen.
Think about it. Elon conjures up a vision of the future where we've managed to increase our solar cell manufacturing capacity by two whole orders of magnitude and have the space launch capability for all of it along with tons and tons of other stuff and the best he comes up with is...GPUs in orbit?
This is essentially the superhero gadget technology problem, where comic books and movies gloss over the the civilization changing implications of some technology the hero invents to punch bad guys harder. Don't get me wrong, the idea of orbiting data centers is kind of cool if we can pull it off. But being able to pull if off implies an ability to do a lot more interesting things. The problem is that this is both wildly overambitious and somehow incredibly myopic at the same time.
A lot of great inventions we now take for granted initially came with little motivation other than being able to kill each other more effectively. GPS, radar, jet engines, drones, super glue, microwaves, canned food, computers, even the internet. Contrary to the narrative of the internet being about sharing science, ARPANET was pushed by the DoD as a means of maintaining comms during nuclear war. It was then adopted by universities and research labs and started along the trajectory most are more familiar with.
The tale of computers is even more absurd. The first programmable, electric, and general-purpose digital computer was ENIAC. [1] It was built to... calculate artillery firing tables. I expect in the future that the idea of putting a bunch of solar into space to run GPUs for LLMs will probably seem, at the minimum - quaint, but that doesn't mean the story ends there.
[1] - https://en.wikipedia.org/wiki/ENIAC
That’s not the point of the person you are replying to. They are saying if we somehow come up with the tech that makes harnessing the sun a thing, the best we can still do is put a bunch of GPUs in space? It makes no sense.
>the best we can
oh, we'll sure find a way to weaponize that energy for example - just imagine all those panels simultaneously turning their reflective back in a way to form gigantic mirror to focus reflected solar energy on your enemy, be that enemy in space or on the Earth/Moon/Mars ground. Basically space-scale version of 'death ray scyscrapper' https://www.businessinsider.com/death-ray-skyscraper-is-wrea....
I think the Colossus[1] predated the ENIAC but is still in line with your general theme of doing stuff for the military. In this case it was used for cipher breaking, not firing calculations.
You could argue that it doesn't really count though because it was only turing complete in theory: "A Colossus computer was thus not a fully Turing complete machine. However, University of San Francisco professor Benjamin Wells has shown that if all ten Colossus machines made were rearranged in a specific cluster, then the entire set of computers could have simulated a universal Turing machine, and thus be Turing complete."
[1] https://en.wikipedia.org/wiki/Colossus_computer
> You could argue that it doesn't really count though because it was only turing complete in theory
Then you have to also count the Z3 which predates the Colossus by 2 years.
[1] https://en.wikipedia.org/wiki/Z3_(computer)
Yes, but as Ron Perlman famously said in the beginning of Fallout, "War never changes".
I would be more shocked that we eliminated war than if we achieved this version of Elon's future.
It makes sense to think that we will continue to make scientific progress through war and self defense.
Reason being, nothing is more motivating than wanting to survive
I'm starting to wonder if a person like Elon with his... morals... is who we want to be creating a vision for the future.
Not to go heads I win, tails you lose, but even if we go down this path - it's the same story because militaries are investing heavily in LLM stuff, both overtly and covertly. Outside of its obvious uses in modeling, data management, and other such things - there also seems to be a fairly widespread belief, among the powers that be, that if you just say the magic words to somebody, that you can make them believe anything. So hyper-scaling LLM potential has direct military application, same as Starlink and Starship.
I think it's much simpler: smart mass surveillance. With LLMs you can finally read and analyze all messages people send to each other
The digital internet began with the telegraphy network in the early 1800s.
Many, many network protocols were developed and used.
> with the telegraphy network in the early 1800s.
Late 1700 actually, and war was indeed a key motivation for the deployment of the Télégraphe Chappe.
See "The Victorian Internet: The Remarkable Story of the Telegraph and the Nineteenth Century's On-line Pioneers"
https://www.amazon.com/dp/162040592X
Télégraphe Chappe was a semaphore system using flags. It was not an electrical telegraph, nor was it binary.
Really? That is so interesting - which ones? Any ancestors of commonly used ones today?
Off the top of my head BIX, Prodigy, Compuserve, MCIMail, BBS, Ethernet, Token Ring, $25 Network, AOL, Timeshare, Kermit, Fax
Anyone with 2+ computers immediately thought about connecting them.
Yeah it does not make a whole lot of sense as the useful lifespan of the gpus in 4-6 years. Sooo what happens when you need to upgrade or repair?
This is a question that analysts don't even ask on earnings calls for companies with lowly earthbound datacenters full of the same GPUs.
The stock moves based on the same promise that's already unchecked without this new "in space" suffix:
We'll build datacenters using money we don't have yet, fill them with GPUs we haven't secured or even sourced, power them with infrastructure that can't be built in the promised time, and profit on their inference time over an ever-increasing (on paper) lifespan.
> This is a question that analysts don't even ask
On the contrary, data centers continue to pop up deploying thousands of GPUs specifically because the numbers work out.
The H100 launched at $30k GPU and rented for $2.50/hr. It's been 3 years since launch, the rent price is still around $2.50.
During these 3 years, it has brought in $65k in revenue.
Beyond GPUs themselves, you also have other costs such as data centers, servers and networking, electricity, staff and interest payments.
I think building and operating data center infrastructure is a high risk, low margin business.
They can run these things at 100% utilization for 3 years straight? And not burn them out? That's impressive.
I don't see anything impressive here?
Not really. GPUs are stateless so your bounded lifetime regardless of how much you use them is the lifetime of the shitties capacitor on there (essentially). Modulo a design defect or manufacturing defect, I’d expect a usable lifetime of at least 10 years, well beyond the manufacturer’s desire to support the drivers for it (ie the sw should “fail” first).
> the useful lifespan of the gpus in 4-6 years. Sooo what happens when you need to upgrade or repair?
Average life of starlink satellite is around 4-5 years
damn. at this point its not even about a pretense for progress, just a fetish for a very dirty space
Same that happens with Starlink satellites that are obsolete or exhausted their fuel - they burn up in the atmosphere.
A "fully and rapidly reusable" Starship would bring the cost of launch down orders of magnitude, perhaps to a level where it makes sense to send up satellites to repair/refuel other satellites.
With zero energy cost it will run until it stops working or runs out of fuel, which I'm guessing is between 5-7 years.
5 to 7 months given they want 100kw Per ton and magical mystery sauce shielding is going to do shit all.
> Sooo what happens when you need to upgrade or repair?
The satellite deorbits and you launch the next one.
not to mention that radiation hardening of chips has a big impact on cost and performance
You could immersion cool them and get radiation resistance as a bonus.
Yes, because launching then immersed in something that will greatly increase the launch weight will help...
The data centers in space is 100% about Golden Dome,
https://wikipedia.org/wiki/Golden_Dome_(missile_defense_syst...
Nope, it's 100% about building the stock valuation of SpaceX for an IPO in the face of significant risk from a cold war its CEO started on X with the U.S. federal government and increasing competition from Blue Origin, Quinfan and Guowang. DOD will play Bedrock vs Grok until there is feature parity and then make a decision not based on the features.
Disclaimer: Not an Elon hater, but far from a sycophant, similar to how I felt about Steve Jobs for 40+ years.
Exactly, this is about attaching the AI hype bubble to all of his dealings before he offloads with an IPO (that still leaves him with 75% of the stock).
And that's why the best way to use Superman's powers is in making him turn a giant crank
(yes I fully agree with you!)
So what are the other things? You said he glossed over them and didn't mention a single one.
Reliably and efficiently transport energy generated in space back to earth, for starters
Or let me guess, its going to be profitable to mine crypto in space (thereby solving the problem of transporting the "work" back to earth)
Overview energy has done interesting work in this area.
It's always better to generate electricity on the ground than attempt to beam it to the ground from space. The efficiency loss of beamed power is huge.
The efficiency loss of nighttime is approximately 100% if we’re talking about solar energy. At least at a most basic level, it’s not totally absurd to stick some kind of power beaming contraption in space where it is mostly not shadowed by the Earth and beam power to a ground station.
I concur it’s not necessarily totally absurd — but when you consider that such contraptions require large — very large! — receiving arrays to be built on the ground, it’s hard to avoid concluding that building gigantic photovoltaic arrays in, say Arizona (for the US) along with batteries for overnight buffering and transmission lines would still be massively more efficient.
Is that more or less absurd than making deals with our neighbours to share their electricity? Build some solar farms around the planet and then distribute it over wire.
I honestly don't know the answer. I know there's some efficiency loss running over long wires too but I don't know what's more realistic.
There is absolutely nothing realistic about power transmission from space to earth, wired or wireless.
We have these things called batteries, you charge them during the day, and drain them at night.
A solar+battery setup is already cheaper than a new gas plant. Beaming power from space is absolutely asinine, quite frankly. The losses are absurd, the sun already does it 24/7, and we know how to make wires and batteries to shuffle the sun's power around however we need to. Why on earth would we involve satellites?
Why would you transfer the energy to earth? The energy powers ai compute = $
Dead on, You can transmit data to and from space and have the compute completed at potentially fractions of the cost.
Tell me about your cooling medium in space
A large piece of aluminum with ammonia pumped through it?
Right up to the radiation limit and then you'll either have to throttle your precious GPUs or you'll be melting your satellite or at least the guts of it. You're looking at an absolutely massive radiator here, many times larger than the solar panels that collect the energy to begin with.
Nothing about this is sounding economically competitive with ground based solutions
Where does the heat collected by amminia get evacuated?
Through thermal radiation, it's called radiative cooling.
But it's not trivial indeed, especially if you want good power density in your space data center.
Not sure why this is downvoted. Much cheaper to transfer data than energy.
If we (as in "civilization") were able to produce that many solar panels, we should cover all the deserts with them. It will also shift the local climate balance towards a more habitable ecosystem, enabling first vegetation and then slowly growing the rest of the food chain.
> But being able to pull if off implies an ability to do a lot more interesting things.
Those interesting things won't pump up the perceived value of Musk companies to stratospheric levels - or dare I say - to the moon. He needs the public to believe that to earn the trillion-dollar package from the Tesla-Twitter-SpaceX conglomerate, even if the latter turns out to be the only profitable arm of the conglomerate.
You really can't grasp that GPUs scaled at this level is the most ambitious thing possible? That it will be the foundation of unfathomable technological innovation?
Will it, though?
Perhaps parent was being sarcastic.
Honestly, there's not a lot else I can think of if your goal is find some practical and profitable way to take advantage of relatively cheap access to near-Earth space. Communication is a big one, but Starlink is already doing that.
One of the things space has going for it is abundant cheap energy in the form of solar power. What can you do with megawatts of power in space though? What would you do with it? People have thought about beaming it back to Earth, but you'd take a big efficiency hit.
AI training needs lots of power, and it's not latency sensitive. That makes it a good candidate for space-based compute.
I'm willing to believe it's the best low-hanging fruit at the moment. You don't need any major technological advances to build a proof-of-concept. Whether it's possible for this to work well enough that it's actually cheaper than an equivalent terrestrial datacenter now or in the near future is something I can't answer.
You don't need any major technological advances to build a proof-of-concept
You do - cooling those datacenters in space is an unsolved problem.
Sure it is, just not economically at that scale yet. But if Starship brings the cost to orbit down significantly, maybe.
We have radiators on the ISS. Even if you kept the terrible performance of those ancient radiator designs (regularly exposed to sunlight, simplistic ammonia coolant, low temperature) you could just make them bigger and radiate the needed energy. Yes it would require a bit of engineering but to call it an "unsolved problem" is just exaggerating.
It's a solved problem. The physics is simply such that it's really inefficient.
> ... we'd need a system 12.5 times bigger, i.e., roughly 531 square metres, or about 2.6 times the size of the relevant solar array. This is now going to be a very large satellite, dwarfing the ISS in area, all for the equivalent of three standard server racks on Earth.
https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
The gist of it is that about 99% of cooling on earth works by cold air molecules (or water) bumping into hot ones, and transferring heat. There's no air in space, so you need a radiator 99x larger than you would down here. That adds up real fast.
That’s the secret plan - cover LEO with solar cells and radiators, limiting sunlight on the ground, rendering ground base solar ineffective, cool earth and create more demand for heating; then sell expensive space electricity at a huge premium. Genius!
Bezos has been pushing manufacturing-in-space for a long time, as a ideal candidate for what to do in space that you might prefer to not do on Earth. Robotics, AI automation, manufacturing - combo it in space, let the robots manufacture for us in space. Abundant energy, low concerns about most forms of pollution. We'll need to dramatically improve our ability to transit mass to and from cheaply first of course (we're obviously talking many decades into the future).
That is a fun thought experiment, as we wouldn't want to manufacture too far away from earth we may still be within the earth's atmosphere. I wonder what effect dumping greenhouse gases into the very upper levels of the atmosphere would have in comparison to doing it lower down. My assumption is it would eventually sink to a lower density layer, having more or less the same impact.
When Bezos first mentioned drone delivery, many intelligent, serious people laughed at it and accused of Bezos running out of ideas as Amazon was stagnant
> Bezos has been pushing manufacturing-in-space for a long time, as a ideal candidate for what to do in space that you might prefer to not do on Earth. Robotics, AI automation, manufacturing - combo it in space, let the robots manufacture for us in space.
LOL, this seems so far off from the reality of what manufacturing looks like in reality. - sending raw materials up there - service technicians are necessary ALL THE TIME, in fully automated production lines - sending stuff back down
Maybe I lack vision, but data centers in space is a 1000x times better idea and that is already a terrible idea.
Well you see, what you do is send a bunch of humanoid robots up there to do all the work.
(please don't ask what we do when those break down)
>>sending raw materials up there
That's what asteroid mining is for.
>>service technicians are necessary ALL THE TIME
Optimus is already very well tele-operated. Even though over time it can likely be trained to do specific tasks far better than even humans.
The show For All Mankind kind-of hinted at how the labor problem would be solved: recruit like the military and promise huge bonuses that will probably not be realized because space is risky business
I think it makes more sense if you invert the manufacturing cycle.
Automated asteroid mining, and asteroid harvesting, are potential areas where we have strong tech, a reasonable pure automation story, and huge financial upsides. Trillion dollar asteroids... If we’re sourcing metals out there, and producing for orbital operations or interplanetary shenanigans, the need for computing and automation up there emerges.
And I imagine for the billionaire investor class now is the window to make those kinds of plays. A whole set of galactic robber barons is gonna be crowned, and orbital automation is critical to deciding who that is.
Hate to say this, but manufacturing bitcoin would make the most sense. And hard to see how even that would work.
But everyone is crazy about GPU’s right now. Why not ride that wave for extra investment? All the benefits transfer to all the other things we can do with it.
We also shouldn't overlook the benefits we might realize if humanity achieved the incredible engineering and technical capacity necessary to make this version of porcine flight happen.
This is such a hypebeast paragraph.
Datacenters in space are a TERRIBLE idea.
Figure out how to get rid of the waste heat and get back to me.
That's not a new problem that no one has dealt with before. The ISS for instance has its External Active Thermal Control System (EACTS).
It's not so much a matter of whether it's an unsolvable problem but more like, how expensive is it to solve this problem, what are its limitations, and does the project still makes economic sense once you factor all that in?
It's worth noting that the EACTS can at maximum dissipate 70kW of waste heat. And EEACTS (the original heat exchange system) can only dissipate another 14kW.
That is together less than a single AI inference rack.
And to achieve that the EACTS needs 6 radiator ORUs each spanning 23 meters by 11 meters and with a mass of 1100 kg. So that's 1500 square meters and 6 and a half metric tons before you factor in any of the actual refrigerant, pumps, support beams, valve assemblies, rotary joints, or cold side heat exchangers all of which will probably together double the mass you need to put in orbit.
There is no situation where that makes sense.
-----------
Manufacturing in space makes sense (all kinds of techniques are theoretically easier in zero G and hard vacuum).
Mining asteroids, etc makes sense.
Datacenters in space for people on earth? That's just stupid.
I'm a total noob on this.
I get that vacuum is a really good insulator, which is why we use it to insulate our drinks bottles. So disposing of the heat is a problem.
Can't we use it, though? Like, I dunno, to take a really stupid example: boil water and run a turbine with the waste heat? Convert some of it back to electricity?
What do you do with the steam afterwards? If you eject it, you have to bring lots of it with your spacecraft, and that costs serious money. If you let it condensate to get water again, all you did is moving some heat inside the spacecraft, almost certainly creating even more heat when doing that.
It's a good question, but in a closed system (like you have in space) the heat from the turbine loop has to go somewhere in order to make it useful. Let's say you have a coolant loop for the gpus (maybe glycol). You take the hot glycol, run it through your heat exchanger and heat up your cool, pressurized ammonia. The ammonia gets hot (and now the glycol is cool, send it back). You then take the ammonia and send it through the turbine and it evaporates as it expands and loses pressure to spin the turbine. But now what? You have warm, vaporized, low pressure ammonia, and now you need to cool it down to start over. Once it's cool you can pressurize it again so you can heat it up to use again, but you have to cool it, and that's the crux of the issue.
The problem is essentially that everything you do releases waste heat, so you either reject it, or everything continues to heat up until something breaks. Developing useful work from that heat only helps if it helps reject it, but it's more efficient to reject it immediately.
A better, more direct way to think about this might be to look at the Seebeck effect. If you have a giant radiator, you could put a Peltier module between it and you GPU cooling loop and generate a little electricity, but that would necessarily also create some waste heat, so you're better off cooling the GPU directly.
Harder to direct waste heat in space if you dont have gravity for convection.
The ISS consumes roughly 90kW. That’s about *one* modern AI/ML server rack. To do that they need 1000 m^2 of radiator panels (EACTS). So that’s the math: every rack needs another square kilometer of stuff put into orbit. Doesn’t make sense to me.
1000m2 is not a square kilometer (1 square kilometer is 1mil m2)
And what happens every time a rack (or node) fails? Does someone go out and try to fix it? Do we just "deorbit" it? How many tons per second of crap would we be burning in the upper atmosphere now? What are the consequences of that?
How do the racks (or nodes) talk to eachother? Radios? Lasers?
What about the Kessler Syndrome?
Not a rocket scientist but 100% agree this sounds like a dead end.
Communication is a well-understood problem, and SpaceX already has Starlink. They might need pretty high bandwidth, but that's not necessarily much of a problem in space. Latency could be a problem, except that AI training isn't the sort of problem where you care about latency.
I'd be curious where exactly they plan to put these datacenters... In low Earth orbit they would eventually reenter, which makes them a pollution source and you'd have no solar power half the time.
Parking them at the Earth-Sun L1 point would be better for solar power, but it would be more expensive to get stuff there.
> SpaceX already has Starlink. They might need pretty high bandwidth
you mean the network that has less capacity than a fibre pair per coverage area?
> you'd have no solar power half the time
Polar orbit.
Seasons mess that up unless you're burning fuel to make minor plane changes every day. Otherwise you have an equinox where your plane faces the sun (equivalent to an equatorial orbit) and a solstice where your plane is parallel to the sun (the ideal case).
https://en.wikipedia.org/wiki/Sun-synchronous_orbit A Sun Synchronous orbit at the Day-Night terminator solves this issue
True. It would a tradeoff with the fuel consumed vs doubling power output.
It makes sense to target a higher operating temperature, like 375K. At some point, the energy budget would reach an equilibrium. The Earth constantly absorbs solar energy and also dissipates the heat only by radiative cooling. But the equilibrium temperature of the Earth is still kind of cool.
I guess the trick lies in the operating temperature and the geometry of the satellites.
Asking for a friend (who sucks at thermodynamics:) could you use a heat pump to cool down the cold end more and heat up the hot end much higher? Heat radiation works better the higher the temperature?
Not sure about the effectiveness of a heat pump in this use case.
>Heat radiation works better the higher the temperature?
The power output is proportional to T^4 according to the Stefan-Boltzmann law.
I agree that data centers in space is nuts.
But I think there's solutions to the waste heat issue
https://www.nasa.gov/centers-and-facilities/goddard/engineer...
The distinction is that what they are doing for Webb is trying to dissipate small amounts of heat that would warm up sensors past cryogenic temperatures.
Like on the order of tens or hundreds of watts but -100C.
Dissipating heat for an AI datacenter is a different game. A single AI inference or training rack is going to be putting out somewhere around 100kW of waste heat. Temps don't have to be cryogenic but it's the difference between chiselling a marble or jade statue and excavating a quarry.
That's a solution for minuscule amounts of heat that nevertheless disturb extremely sensitive scientific experiments. Using gold, no less. This does not scale to a crapton of GPU waste heat.
Just have to size radiators correctly. Not a physics problem. Just an economic one.
Main physics problem is actually that the math works better at higher GPU temps for efficiency reasons and that might have reliability trade off.
Anything is possible here, it's just there's no goddamn reason to do any of this. You're giving up the easiest means of cooling for no benefit and you add other big downsides.
It's scifi nonsense for no purpose other than to sound cool.
It's about creating a flywheel for scale.
Getting better at creating and erecting solar panels & AI datacenters on earth is all well and good, but it doesn't advance SpaceX or humanity very much. At lot of the bottlenecks there are around moving physical mass and paperwork.
Whereas combining SpaceX & xAI together means the margins for AI are used to force the economies of scale which drives the manufacturing efficiencies needed to drive down launch etc.
Which opens up new markets like Mars etc.
It is also pushing their competitive advantage. It leaves a massive moat which makes it very hard for competitors. If xAI ends up with a lower cost of capital (big if - like Amazon this might take 20 years horizon to realize) but it would give them a massive moat to be vertically integrated. OpenAI and others would be priced out.
If xAI wants to double AI capacity then it's a purely an automation of manufacturing problem which plays to Elons strengths (Tesla & automation). For anyone on earth doubling capacity means working with electricity restrictions, licensing, bureaucracy, etc. For example all turbines needed for electricity plants are sold years in advance. You can't get a new thermal plant built & online within 5 years even if you had infinite money as turbines are highly complex and just not available.
Hmm, Elon really did run that flywheel pretty well. He built the Roadster to drum up some cash and excitement so he could develop the Model S, then he used that success to do the Model X, and then he expanded capacity to develop the 3 and Y, and he reinvested the profits to develop the Model 2, finally bringing EVs to the masses, displacing ICEs everywhere, and becoming the undisputed leader of both EV and battery manufacturering.
Oh wait, that didn’t actually happen, because he got distracted or something? He doesn’t really have battery capacity worth writing home about, the Chinese are surpassing Tesla in EV manufacturing, and Waymo is far ahead in self-driving.
The amazing space computation cost reduction process sounds rather more challenging than the Model 2, and I’m not sure why anyone should bet on Elon pulling it off.
We must be living in parallel universes.
Tesla invested into the first Lotus roadster - and put that cash into the S then the X. Used that cash to build the worlds largest factories and make the 3 & Y which sold at enormous volumes - so large in fact that the S & X are now tiny single percentages of sales which is why Tesla is stopping manufacturing them now.
Tesla is one of the very few vehicle manufactures which makes a profit manufacturing vehicles. Tesla throws off cash which allows the flywheel to keep spinning.
Tesla is now operating fully autonomous rides. They've constantly proved their naysayers wrong at every turn in time. What the Chinese are doing in battery tech is irrelevant to US vehicles as they will never be allowed to sell in the US which is Teslas largest market.
The model 2 has the possibility of being profitable at insanely low purchase price which has the potential to completely disrupt the economics of US sales in such a way that legacy auto could well be bankrupt in 5-10 years. Who will be making Waymo's vehicles then?
Hey remember that time someone had their Tesla running down the highway and the superior self-driving capability failed to see an 18 wheeler that crossed the road and the person was decapitated and there are videos of that complete with blood spray?
There is nothing we need on Mars other than science. It's not a market because there isn't money to be made outside of what is required to do whatever economically useless but scientifically valuable efforts we can convince people to fund.
We can't build an independent colony we can't live there any time soon. Arguably it may never make sense to live there.
With that attitude mankind would still be living in caves. Why build a farm and stay in one place - we should follow the animals around.
1. Mankind never systematically lived in caves; that's just where remains and rock paintings are more likely to have survived.
2. Farming didn't evolve from a vision of "let's stay in one place, so let's find a way to do it"; it evolved from the gradual application of accumulated practical knowledge under real constraints until eventually it was possible to stay in one place. If Paleoelon had somehow convinced early humanity to abandon hunter-gathering and settle into a sedentary life because he had a vision for new markets around farming it would have led to the earliest famine.
All right, so how is it that all you geniuses out here are totally right about this, but all the dullards at SpaceX and XAI, who have accomplished nothing compared to you lot, are somehow wrong about what they do every day?
I know being right without responsibility feels amazing but results are a brutal filter.
I once had a job mopping floors and was quite successful at it, even if I say so myself. Based on my experience, do you think it is reasonable for me to claim that I will eventually develop techniques for cleaning the oceans of all plastic waste? Folks are criticizing the pie in the sky claims, not that they can do anything at all.
Seems a bit of both. But no disparagement to your floor mopping (as I once was a dishwasher in a commercial kitchen myself), but there's a big gap between cleaning a floor, or a dish, and creating frontier models and spaceships.
That said: I think solar is niche, and a moon-shot for how they want it. Nuclear is the future of reliable energy for human civilization.
I think the K-scale is the wrong metric. I don't think we should be trying to take all the sun's energy as a goal (don't blot out the sun! don't hide it in a bushel!), or as a civilizational utiltiy - I'm sure better power supplies will come along.
Data centers ultimately need to provide power and remove heat. Solar might be a little easier for power in space, maybe, but heat is an absolute no-go, stop, this will never ever work. You can't engineer your way out of the fact that space is a vacuum.
This vision doesn't come from those great engineers, but from Elon, the guy who promised Hyperloop, FSD in 2 years 10 years ago, and lots of other BS
spacex is one thing but xai accomplished what? the most racist csam prone llm?
I'm not aware of this - What's that?
Probably shouldn't speak to the brilliance of xAI engineers when you've never heard of their work
Is whatever that is their work?
There's no reason to think the brilliant minds at SpaceX are supportive of focusing their mission in any manner-what-so-ever on datacenters in space. You can't call on their genius as the supportive argument accordingly.
I disagree, I think the idea of a cabal of reactionary comrades inside SpaceX is activist fantasy. I think SpaceX only does what it does with full committment of its people: mind, body, spirit.
I think there's a scenario where that's true: one where the head of your company is collaborative and deferential to expertise.
There's another scenario, though: one where the head of your company is a bull in a China shop, whose successes have come almost exclusively through a Barnum-esque scheme of cascading bravado and marketing genius without much expertise, but a marvelous ability to sell any idea purely via unearned gravitas.
The former is less sexy: I've compiled loads of talented people, and we're going to solve very hard problems, even some that seem impossible.
The latter is very sexy: I'm a genius and we're going to accomplish the impossible in one year via sheer force of my grand will. And even if it doesn't actually happen, I'll sell you on the next vision.
It seems like you’re ascribing to Elon some kind of magic, where you feel he’s breaking the rules of what should be allowed in order to achieve success. Is it impossible you simply don’t understand how what he does works?
This place has derangement syndrome unfortunately. Such pessimists, it’s a bit sad
Only people who never interacted with data center reliability think it's doable to maintain servers with no human intervention.
Microsoft did do the experiment (Project Natick) where they had "datacenters" in pods under the sea with really good results. The idea was simply to ship enough extra capacity, but due to the environment, the failure rates where 1/8th of normal.
Still, dropping a pod into the sea makes more sense than launching it into space. At least cooling, power, connectivity and eventual maintenance is simpler.
The whole thing makes no sense and is seems like it's just Musk doing financial manipulation again.
https://news.microsoft.com/source/features/sustainability/pr...
Whoa there, space-faring sysadmin. You really want that off-world contract tho?
Haha, hard pass on the job. I prefer my oxygen at 1 atm.
I'm not a data center technician myself, but I have deep respect for those folks and the complexity they manage. It's quite surprising the market still buys Musk's claims day after day.
There are a class of people who may seem smart until they start talking about a subject you know about. Hank Green is a great example of this.
For many on HN, Elon buying Twitter was a wake up call because he suddenly started talking about software and servers and data centers and reliability and a ton of people with experience with those things were like "oh... this guy's an idiot".
Data centers in space are exactly like this. Your comment (correctly) alludes to this.
Companies like Google, Meta, Amazon and Microsoft all have so many servers that parts are failing constantly. They fail so often on large scales that it's expected things like a hard drive will fail while a single job might be running.
So all of these companies build systems to detect failures, disable running on that node until it's fixed, alerting someone to what the problem is and then bringing the node back online once the problem it's addressed. Everything will fail. Hard drives, RAM, CPUs, GPUs, SSDs, power supplies, fans, NICs, cables, etc.
So all data centers will have a number of technicians who are constantly fixing problems. IIRC Google's ratio tended to be about 10,000 servers per technician. Good technicians could handle higher ratios. When a node goes offline it's not clear why. Techs would take known good parts and basically replacce all of them and then figure out what the problem is later, dispose of any bad parts and put tested good parts into the pool of known good parts for a later incident.
Data centers in space lose all of this ability. So if you have a large number of orbital servers, they're going to be failing constantly with no ability to fix them. You can really only deorbit them and replace them and that gets real expensive.
Electronics and chips on satellites also aren't consumer grade. They're not even enterprise grade. They're orders of magnitude more reliable than that because they have to deal with error correction terrestial components don't due to cosmic rays and the solar wind. That's why they're a fraction of the power of something you can buy from Amazon but they cost 1000x as much. Because they need to last years and not fail, something no home computer or data center server has to deal with.
Put it this way, a hardened satellite or probe CPU is like paying $1 million for a Raspberry Pi.
And anybody who has dealt with data centers knows this.
Great comment on hardware and maintenance costs, and in comparison Elon wrote "My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space." It's a pity this reads like the entire acquisition of xAi is based on "Elon's napkin math" (maybe he checked it with Grok)
The deal they made values xAI at $230 Billion. It’s a made up number, with no trustworthy financial justification to back it up. It is set to provide a certain return to xAI’s investors (the valuation decides the amount you get per share), who in turn are bailing out the earlier acquisition of X (Twitter). All of this is basically a shell game where Elon is using one company to bail out another. It’s a way of reducing the risk of new ventures by spreading them out between his companies. It’s also really bad for SpaceX employees and investors, who are basically subsidizing other companies.
The thing is, everyone knows Elon is not a real CEO of any of these companies. There isn’t enough time to even be the CEO of one company and a parent. This guy has 10 companies and 10 children. He’s just holding the position and preventing others from being in that position, so he can enact changes like this. And his boards are all stacked with family members, close friends, and sycophants who won’t oppose his agenda.
As both are private companies none of this matters if the investors of both companies are happy.
Most of the investors don’t even have a choice. Nor do all the other shareholders like employees. And the boards of Musk companies are stacked with his yes men.
Ah yes, my favourite kind of engineering: financial engineering
He's bailing out one of his failing ventures with one of his so far successful ones. The BS napkin math isn't the reason he's doing it. It's the excuse for doing it.
Or he's having another mental break because he knocked up yet another woman and is going to have yet another kid he can't remember the name of.
Can you provide a link for that quote, because that quote is absolute stupidity.
It's in the article that you're commenting on, https://www.spacex.com/updates#xai-joins-spacex.
Oh, ffs.
Haha. It's less than 1,000 words that would take less than 5 minutes to read.
I bet much less than half of the hundreds of HN commenters here bother to read it. Many are clearly unfamiliar with its content.
I can't, I don't want it in my head :/
Thanks for putting words to that; the paragraph which most stuck out to me as outlandish is (emphasis mine):
I'm deeply disillusioned to arrive at this conclusion but the Occam's Razor in me feels this whole acquisition is more likely a play to increase the perceptual value of SpaceX before a planned IPO."what if we move all our data center needs into my imagination, things are running so much smoother there"
> but they cost 1000x as much
Compute power has increased more than 1000x while the cost came down.
I recall paying $3000 for my first IBM PC.
> they need to last years and not fail
Not if they are cheap enough to build and launch. Quantity has a quality all its own.
Have you heard of cosmic radiation?
Cosmic rays take time to destroy them.
First of all Twitter had basically no downtime since he bought it, so all the 'internet experts' posting their thoughts were completely dead wrong. If anything Twitter was far more reliable than Microsoft has been these past few years.
You are assuming things need to run the same way in space, for instance you mentioned fans, you won't have any in space. You also won't have any air, dust, static, or any moving parts.
You are assuming the costs to launch to orbit are high, when the entire point of Spacex's latest ship is to bring the cost to launch so low that it is cheaper per ton than an airplane flight.
Maintenance would be nice but you are saying this like Elon Musk's company doesn't already manage the most powerful datacenters on the planet.
You have no clue what you are talking about regarding cosmic rays and solar wind, these will literally be solar powered and behind panels and shielding 100% of the time.
I went looking through your comments. 75% of them (and probably 90% in the lasst 2 years) were Elon related. Tesla, SpaceX, Grok, Twitter, DOGE, etc. Quite a lot of comments for 101 karma if I'm being real.
Why do you feel this kneejerk reaction to defend Elon and his companies? You'll never be him. He doesn't care about you. He'd use you for reactor shielding for an uptick in Tesla share price without a second's hesitation. This is cultish behavior.
Do you have any idea who you're defending? I'll give you just one example. A right-wing influencer named Dom Lucre uploaded CSAM to Twitter, a video. But he didn't just upload it. He watermarked it first so had it on his computer and then postporcessed it. It was I believe up for days. This was apparently a video so bad that mere possession should land you in prison. And the fact that the FBI didn't arrest him basically tells you he'd an FBI asset. After taking days to ban him, Elon personally intervened to unban him. Why? Because reasons.
And this is the same man who it's becoming clear was deeply linked with Jeffrey Epstein, as was his brother [1].
Bringing this back to the original point: this is why Twitter lost 80% of its value after Elon acquired it. Advertisers fled because it became a shithole for CSAM and Nazis.
As for "basically no downtime" that's hilarious. I even found you commenting the classic anecdote "it was fine for me" (paraphrased) on one such incident when Twitter DDOSed itself [2].
Your cultish devotion here is pretty obvious eg [3]. I'm genuinely asking: what do you get out of all this?
[1]: https://www.axios.com/local/boulder/2026/02/02/kimbal-musk-j...
[2]: https://news.ycombinator.com/item?id=36555897
[3]: https://news.ycombinator.com/item?id=42836560
Might be why he's also investing in building their own fabs - if he can keep the silicon costs low then that flips a lot of the math here.
But … but what if we had solar-powered AI SREs to fix the solar-powered AI satellites… /in space/?
Maintaining modern accelerators requires frequent hands-on intervention -- replacing hardware, reseating chips, and checking cable integrity.
Because these platforms are experimental and rapidly evolving, they aren't 'space-ready.' Space-grade hardware must be 'rad-hardened' and proven over years of testing.
By the time an accelerator is reliable enough for orbit, it’s several generations obsolete, making it nearly impossible to compete or turn a profit against ground-based clusters.
On the other hand, Tesla vehicles have similar hardware built into them, and don't require such hands-on intervention. (And that's the hardware that will be going up.)
Car-grade inference hardware is fundamentally different from data center-grade inference hardware, let alone the specialized, interconnected hardware used for training (like NVLink or complex optical fabrics). These are different beasts in terms of power density, thermal stress, and signaling sensitivity.
Beyond that, we don't actually know the failure rate of the Tesla fleet. I’ve never had a personal computer fail from use in my life, but that’s just anecdotal and holds no weight against the law of large numbers. When you operate at the scale of a massive cluster, "one-in-a-million" failures become a daily statistical certainty.
Claiming that because you don't personally see cars failing on the side of the road means they require zero intervention actually proves my original point: people who haven't managed data center reliability underestimate the sheer volume of "rare" failures that occur at scale.
https://x.com/elonmusk/status/2017792776415682639
For what it's worth, this project plans to use Tesla AI5/AI6 hardware for the first launches.
Not only the sibling comments points, but cars aren't exposed to the radiation of space...
Do they need to be maintained? If one compute node breaks, you just turn it off and don't worry about it. You just assume you'll have some amount of unrecoverable errors and build that into the cost/benefit analysis. As long as failures are in line with projections, it's baked in as a cost of doing business.
The idea itself may be sound, though that's unrelated to the question of whether Elon Musk can be relied on to be honest with investors about what their real failure projections and cost estimates are and whether it actually makes financial sense to do this now or in the near future.
AI clusters are heavily interconnected, the blast radius for single component failure is much larger than running single nodes -- you would fragment it beyond recovery to be able to use it meaningfully.
I can't get in detail about real numbers but it's not doable with current hardware by a large margin.
This guy invented reusable rockets that land themselves. I'm sure xAI is not just one guy. Plenty of talented people work there.
I couldn't believe that was an actual quote from the article. It is.
These people are legit insane.
Smart people call it "story telling" /s. Musk bullshit and constant lie made him the richest person in the world. No reason not to continue.
Not insane at all. They are perfectly sane and know words can be twisted to justify just about anything, when stating the actual goals is unsavory.
No it's just Musk's Big Idea for spacex to hype it before IPO. It's their version of FSD, robots etc.
You've got to hand it to him, he is a bullshitter par excellence.
How people still believe his bullshit is unfathomable.
You don't have to believe one's bullshit. You just have to believe others will believe the bullshit.
Yeah, I remember people saying that about making 1m model 3s per year, landing rockets, getting 10k+ satellite privately into orbit, and getting millions of subscribers using internet via those satellites.
Maybe just maybe the guy does actually get things done, and if you didn't hate him you'd see that?
(yes, there are some things he hasn't gotten done. That doesn't take away from what he has gotten done)
Context missing. This is in reference to a vision the (distant?) future where the satellites are manufactured in factories on the Moon and sent into space with mass drivers.
Full paragraph quote comes from:
> While launching AI satellites from Earth is the immediate focus, Starship’s capabilities will also enable operations on other worlds. Thanks to advancements like in-space propellant transfer, Starship will be capable of landing massive amounts of cargo on the Moon. Once there, it will be possible to establish a permanent presence for scientific and manufacturing pursuits. Factories on the Moon can take advantage of lunar resources to manufacture satellites and deploy them further into space. By using an electromagnetic mass driver and lunar manufacturing, it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power. >
> This is in reference to a vision the (distant?) future where the satellites are manufactured in factories on the Moon and sent into space with mass drivers.
In the meantime, how about affordable insulin for everybody?
Why is it cheaper to ship all of the materials to space, then to the moon for assembly (which also includes shipping all of the people and supplies to keep them alive), then back into space vs just…
building them on earth and then shipping them up?
We’re not exactly at a loss for land over here.
https://en.wikipedia.org/wiki/Lunar_resources
In situ manufacturing. You just have to send enough to build the thing that builds the factory.
It's not like satellites need anything like computer chips, which are finicky things to build that require parts with a sole supplier on the entire planet.
You can make propellant on the Moon (aluminum based solid fuels), and the energy to get into orbit or into deep space is far, far less that from Earth’s surface.
Why would satellites be manufactured on the moon? There's nothing on the moon. The raw materials would have to be ferried over first. What would be the point?
It would appeal to naive technofetishists, the same crowd of investors enamored by many of Elon's other impossible schemes.
The moon mfg makes significantly more sense than the hilarious plan to establish a permanent Mars base in the next 50 years, but that's not saying much.
> Why would satellites be manufactured on the moon? There's nothing on the moon. The raw materials would have to be ferried over first. What would be the point?
From lunar regolith you would extract: oxygen, iron, aluminum, titanium, silicon, calcium, and magnesium.
From the poles you can get fuel (water ice -> water + hydrogen + oxygen).
The real constraint is not materials, but rather power generation, automation reliability, and initial capital investment.
So you have to shuttle machines, energy systems, and electronics.
The moon can supply mass, oxygen, fuel, and structure.
Satellites that would benefit most are: huge comms platforms, space-based power satellites, large radar arrays, deep-space telescopes, etc.
>From lunar regolith you would extract: oxygen, iron, aluminum, titanium, silicon, calcium, and magnesium.
Do we actually know how to do that?
>From the poles
From the poles! So the proposal includes building a planetary-scale railway network on bumpy lunar terrain.
>The moon can supply mass, oxygen, fuel, and structure.
None of those are things we are hurting for down here, though.
> So the proposal includes building a planetary-scale railway network on bumpy lunar terrain.
And that’s from a fascist who barely managed to dig ONE small one lane tunnel under Las Vegas and called it a revolution.
I’m sorry to be rude but people who are still giving musk any credit are stupid at this point.
Oh boy, IA data centers in space. It’s not only ridiculous, but it’s also boring and not even exciting at all.
Power would almost certainly mostly come from solar panels. The SpaceX-xAI press release mentions using mass drivers which are electrically powered. Could make Hydrogen-Oxygen rocket fuel but not needed in Moon's lower gravity/thin atmosphere.
> The real constraint is not materials
It's solvents, lubricants, cooling, and all the other boring industrial components and feedstocks that people seem to forget exist. Just because raw materials exist in lunar regolith doesn't mean much if you can't actually smelt and refine it into useful forms.
> We currently make around 1 TW of photovoltaic cells per year, globally.
Doubling every three years; at that rate it would take about 30 years for 1TW to become 1000TW. Whether on not the trend continues largely depends on demand, but as of right now humanity seems to have an insatiable demand for power.
Space to put them, terrestrially, is not infinite. Demand has a hard ceiling.
I think it largely depends on what bottlenecks exist that we haven’t hit yet.
We’re not going to use 100% of our solar panel manufacturing capacity to power space data centers, specifically because everyone else on the ground is so power-hungry. If we’re being generous, it could maybe top out at 1%, which adds another ~20 years to your timeline for a total of 50. I think it’s safe to say this part is bunk (along with everything else about this plan which is also bunk).
We seem to be using 100% of our DRAM manufacturing for AI. So it's not completely out of the question.
I wonder what the plan is to recycle those. Without a plan to safely bring back all this hardware and recycling it, we'll deplete earth from it's mineral. The matter used to build things on earth stays within earth's ecosystem.
Moving matter out continusously at industrial scale with no plan to bring 100% of it back in the ecosystem other than burning it seems quite unsustainable and irresponsable.
Whilst I agree that this glosses over a huge number of technical obstacles, space based solar power could scale more easily than that on earth. Lack of variable weather and gravity means rather than using photovoltaic cells, you can just set up paper thin huge mirrors to focus light and generate steam.
Caveat: my understanding of this largely comes from the book The High Frontier, which is really old and probably inaccurate. I can't think of a reason why this particular point would be wrong though.
Earth does have plenty of sand and iron. Literally all you have to do is grow the sand into a crystal, slice it up, etch some patterns onto it, then add some metal.
Making only 1TW of pv cells per year is a skill issue.
Sure, and copper, and aluminium
In fairness, solar cells can be about 5x more efficient in space (irradiance, uptime).
The quoted "1 TW of photovoltaic cells per year, globally" is the peak output, not the average output. They're only about 20% higher peak output in space… well, if you can keep them cool at least.
But there are no clouds in space and with the right orbit they are always facing the sun
You know how people sometimes dismiss PV by saying "what happens at night or in cloudy weather?"?
Well, what happens over the course of a year of night and clouds is that 1 TW-peak becomes an average of about 110 to 160 GW.
We're making ~1 TW-peak per year of PV right now.
The 1TW is the rated peak power output. It's essentially the same in space. The thing that changes is the average fraction of this sustained over time (due to day/night/seasons/atmosphere, or the lack of all of the above).
It's still the same 1TW theoretical peak in space, it's just that you can actually use close to that full capacity all the time, whereas on earth you'd need to over-provision substantially and add storage, so 1TW of panels can only drive perhaps a few hundred GW of average load.
> the whole capacity
Wouldn’t something like half of the panels be in shadow at any time?
Depends where you put them. The current vogue option is a sun-synchronous orbit: https://en.wikipedia.org/wiki/Sun-synchronous_orbit
polar orbit
It is more than 5x less expensive to get surface area on earth’s surface.
The dominant factor is "balance of system" aka soft costs, which are well over 50%.[0]
Orbit gets you the advantage of 1/5th the PV and no large daily smoothing battery, but also no on-site installation cost, no grid interconnect fees, no custom engineering drawings, no environmental permitting fees, no grid of concrete footers, no heavy steel frames to resist wind and snow loads. The "on-site installation" is just the panels unfolding, and during launch they're compact so the support structure can be relatively lightweight.
When you cost building the datacenter alone, it's cheaper on earth. When you cost building the solar + batteries + datacenter, it (can be) cheaper in space, if you build it right and have cheap orbital launch.
[0] https://en.wikipedia.org/wiki/Balance_of_system
Funny, I would have included transportation as part of the installation cost. You didn't mention that one.
I do say it's predicated on cheap orbital launch. Clearly they expect Starship to deliver, and they're "skating to where the puck will be" on overall system cost per unit of compute.
But yeah, I didn't include that delivering all that stuff by truck (including all the personnel) to a terrestrial PV site isn't free either.
Yeah, soft costs like permitting and inspections are supposedly the main reason US residential solar costs $3/watt while Australian residential solar costs $1/watt. It was definitely the worst and least efficient part of our solar install, everything else was pretty straightforward. Also, running a pretty sizable array at our house, the seasonal variation is huge, and seasonal battery storage isn’t really a thing.
Besides making PV much more consistent, the main thing this seems to avoid is just the red tape around developing at huge scale, and basically being totally sovereign, which seems like it might be more important as tensions around this stuff ramp up. There’s clearly a backlash brewing against terrestrial data centers driving up utility bills, at least on the East Coast of the US.
The more I think about it, the more this seems like maybe not a terrible idea.
Do you imagine there'd be less red tape involved in launching multiple rockets per day carrying heavy payloads?
Like this argument just gets absurd: you're claiming building a data center on earth will be harder from a permitting perspective than FAA flight approval for multiple heavy lift rocket launch and landing cycles.
Mining companies routinely open and close enormous surface area mines all over the world and manage permitting for that just fine.
There's plenty of land no one will care if your build anything on, and being remote with maybe poor access roads is still going to be enormously cheaper then launching a state of the art heavy lift rocket which doesn't actually exist yet.
No maintenance either
Right now it is.
However, the amount of available land is fixed and the demand for its use is growing. Solar isn't the only buyer in this real estate market.
We have so much excess land with no real use for it that our government actually pays farmers to grow corn on it to burn in cars.
Availability of land for solar production isn't remotely a real problem in the near term.
This is really underselling it tbh. Any land that's growing corn in a developed country is likely top 1% of land on earth. Half of the earth is desert and tundra. Which is still incredibly easier to work with than space because you can ship there with a pickup very cheaply. Maybe when nevada and central australia are wall-to-wall solar panels we can check back on space.
The Technology Connections Youtube channel recently did a great video arguing pretty convincingly that the land used to grow corn for cars would be vastly more efficiently used from an energy perspective if we covered it with solar panels.
This.
I feel like everyone just lost their mind.
You just have to remember, most of these people live in high density regions and have little comprehension about how much surface area humanity truly occupies... And that isn't even accounting for offshore constructs.
Realizing the impracticality of it (and that such approaches often collapse under the infeasibility of it) ... wouldn't it be better to... say... cover the Sahara in solar panels instead? That's gotta be cheaper than shipping them into space.
https://inhabitat.com/worlds-largest-solar-project-sahara-de...
https://www.theguardian.com/business/2009/nov/01/solar-power...
(and a retrospective from 2023 - https://www.ecomena.org/desertec/ )
From an engineering perspective, with today’s costs, yes. But don’t forget the political complications of dealing with all those countries that own the Sahara, that’s going to come at it’s own cost.
So now we get the political complications of dealing with all those countries that own ASAT weapons.
the demand is pretty much fake and AI isn't actually making money, just gobbling investors money
Solar can always just go on the roof...
Fortunately there are no downsides to launching solar cells into space that would offset those gains.
Does that include all the required radiators to vent heat?
and of course, the continuous opposite boost needed to prevent the heat vent from knocking them out of orbit.
I think this is all ridiculous, to be clear, but re: this problem couldn't the radiators in theory be oriented so that they vent in opposite directions and cancel out any thrust that would be generated?
Now do waste heat.
Solar cells have exactly the same power rating on earth as in space surely? What would change is their capacity factor and so energy generation.
Solar modules you can buy for your house usually have quoted power ratings at "max STC" or Standard Testing Conditions, which are based on insolation on Earth's surface.
https://wiki.pvmet.org/index.php?title=Standard_Test_Conditi...
So, a "400W panel" is rated to produce 400W at standard testing conditions.
I'm not sure how relevant that is to the numbers being thrown around in this thread, but thought I'd provide context.
The atmosphere is in the way, and they get pretty dirty on earth. Also it doesn't rain or get cloudy in space
Sure but like, just use even more solar panels? You can probably buy a lot of them for the cost of a satellite.
The cost of putting them up there is a lot more than the cost of the cells
“This scale” is not realistic in terms of demand or even capability. We may as well talk about mining Sagittarius A* for neutrons.
You don't even need a particularly large scale, it's efficient resource utilization.
Humanity has a finite (and too small) capacity for building solar panels. AI requires lots of power already. So the question is, do you want AI to consume X (where X is a pretty big chunk of the pie), or five times X, from that total supply?
Using less PV is great, but only if the total cost ends up cheaper than installing 5X the capacity as terrestrial PV farms, along with daily smoothing batteries.
SpaceX is only skating to where they predict the cost puck will be.
You seem to be ignoring the substantial resource cost of putting them up there.
And in geostationary, the planet hardly ever gets in the way. They get full sun 99.5% of the year.
Boosting to geostationary orbit knocks a big chunk out of your payload capacity. Falcon 9 expendable will do 22 tons to LEO and about 8 tons to GTO.
even at 10% (say putting it on some northen pile of snow) it is still cheaper to put it on earth than launch it
I don't disagree
I'm all for efficiency, but I would think a hailstorm of space junk hits a lot harder than one of ice out on the farm.
Except it doesn't melt like regular hail so when further storms come up you could end being hit by the same hail more than once :\
Satellites can adjust attitude so that the panels are always normal to the incident rays for maximum energy capture. And no weather/dust.
You also don't usually use the same exact kind of panels as terrestrial solar farms. Since you are going to space, you spend the extra money to get the highest possible efficiency in terms of W/kg. Terrestrial usually optimizes for W/$ nameplate capacity LCOE, which also includes installation and other costs.
For one or a few-off expensive satellites that are intended to last 10-20 years, then yes. But in this case the satellites will be more disposable and the game plan is to launch tons of them at the lowest cost per satellite and let the sheer numbers take care of reliability concerns.
It is similar to the biological tradeoff of having a few offspring and investing heavily in their safety and growth vs having thousands off offspring and investing nothing in their safety and growth.
Atmospheric derating brings insolation from about 1.367KW/m2 to about 1.0.
And then there’s that pesky night time and those annoying seasons.
It’s still not even remotely reasonable, but it’s definitely much higher in space.
> And then there’s that pesky night time and those annoying seasons.
The two options there are cluttering up the dawn dusk polar orbit more or going to high earth orbit so that you stay out of the shadow of the earth... and geostationary orbits are also in rather high demand.
Put them super super far away and focus all the energy into one very narrow death laser that we trust the tech company to be careful with.
And how much of that power would be spent on high speed communications with Earth that aren't, you know, a megabit or two per second
I grew up on a rural farm in California with a dial-up connection that significantly hampered my ability to participate in the internet as a teenager. I got Starlink installed at my parents' house about five years ago, and it's resulted in me being able to spend considerably more time at home.
Even with their cheapest home plan, we're getting like 100 Mbps down and maybe 20 to 50 up. So it's just not true at all that you would have connections that are a megabit or two per second.
That's not what I'm suggesting. The post says "deep space". If you're going to try to harvest even a tiny percentage of the sun's energy, you're not doing that in Earth's orbit. The comparison is a webcam feed from Mars.
That's pretty much a solved problem. We've had geostationary constellations for TV broadcast at hundreds of megabytes for decades now, and lasers for sat-to-sat comms seems to be making decent progress as well.
> it is possible to put 500 to 1000 TW/year of AI satellites into deep space, meaningfully ascend the Kardashev scale and harness a non-trivial percentage of the Sun’s power
Which satellites are operating from "deep space"?
Those are for video. AI Chat workflows use a fraction of the data.
That's silly on so many levels.
1. the latency is going to be insane.
2. AI video exists.
3. vLLMa exist and take video and images as input.
4. When a new model checkpoint needs to go up, are we supposed to wait months for it to transfer?
5. A one million token context window is ~4MB. That's a few milliseconds terrestrially. Assuming zero packet loss, that's many seconds
6. You're not using TCP for this because the round trip time is so high. So you can't cancel any jobs if a user disconnects.
7. How do you scale this? How many megabits has anyone actually ever successfully sent per second over the distances in question? We literally don't know how to get a data center worth of throughput to something not in our orbit, let alone more than double digit megabits per second.
Grok doesn’t have video as far as I know. I don’t think it’s so absurd. I don’t know how you scale this. But it seems pretty straightforward.
The intractable problem is heat dissipation. There is to little matter in space to absorb excess heat. You'd need thermal fins bigger than the solar cells. The satellite's mass would be dominated by the solar panels and heat fins such that maybe 1% of the mass would be usable compute. It would be 1000x easier to leave them on the moon and dissipate into the ground and 100000x easier to just keep making them on earth.
and, of course and inter-satellite comms and earth base station links to get the data up and down. Starlink is one thing at just above LEO a few hundred km and 20km apart, but spreading these around 10s of thousands of km and thosands of km apart is another thing
What concerns me are the implications if the Dark Forest Theory is correct.
Not to mention… how do you repair it when components fail, especially sensitive electronics against cosmic radiation
> We currently make around 1 TW of photovoltaic cells per year, globally.
China made 1.8 TW of solar cells in 2025.
The raw materials required to make these are incredibly abundant, we make as much as we need.
you realize the factor of 2 you introduce doesn't meaningfully change the order of magnitude that the previous poster is implying right?
You missed the point.
We can make ten or hundred times the number of solar cells we make right now, we just don't have a reason to. The technology is fairly ancient unless you want to compete on efficiency, and the raw materials abundant.
The limit isn't just about the current capacity or the maximum theoretical capacity, it's also about the maximum speed you can ramp.
>Tomorrow?
Eventually :)
Markets are forward looking, and not really bound to 'tomorrow'.
Do we really need to say (on HN especially) that time-to-market does matter?
Not just for startups either. If you ramp up the Polio vaccine in 1 year vs 10 years, that has a big impact on human wellbeing. The two scenarios are not equivalent outcomes, even though it still happens "eventually."
Speed matters.
Sure, speed matters.
Developing new technology happens to matter more.
I'm sure investors are going to do their own analysis on this and reach their own conclusions, you should try betting against it.
Surely the constraint will be the rate at which you can get them into and installed orbit, not the manufacturing rate
you would need 200 times the number of solar cells. I don't think you appreciate the scale that 200x is, especially when China is already:
1. quite good at making solar cells
2. quite motivated to increase their energy production via solar
The bottleneck is deploying solar physically, not making the cells.
We have increased the manufacturing of pretty much every piece of technology you see in front you by 200x at some point in history. Often in a matter of years.
I agree that part of the bottleneck is deploying solar physically. China is the best in the world in deploying solar panels. They are only managing linear increases in their solar capacity, year over year.
You're not considering some important multipliers. In space you're already getting a substantial immediate boost due to greater solar irradiance - no atmosphere or anything getting in the way of those juicy photons. You can also get 24 hour coverage in space. And finally they mention "deep space" - it's unclear what that means but solar irradiance increases on an inverse square law - get half way to the sun and you're getting another 4x boost in power. I'm sure there's other factors I'm not considering as well - space and solar just go quite well together.
You also have all that heat to dissipate....
This is all based on bad math. The people proposing these things don't even have proper scientific and mathematical training to determine what is achievable.
Is there a credible way to cool a space-based data center on that scale?
See Dyson Sphere: https://en.wikipedia.org/wiki/Dyson_sphere
Dyson's paper was literally written in jest.
What do you mean?
>In an interview with Robert Wright in 2003, Dyson referred to his paper on the search for Dyson spheres as "a little joke" and commented that "you get to be famous only for the things you don't think are serious" [...]
To be fair, he later added this:
>in a later interview with students from The University of Edinburgh in 2018, he referred to the premise of the Dyson sphere as being "correct and uncontroversial".[13] In other interviews, while lamenting the naming of the object, Dyson commented that "the idea was a good one", and referred to his contribution to a paper on disassembling planets as a means of constructing one.
Sources are in: https://en.wikipedia.org/wiki/Dyson_sphere
Thanks for pointing out those follow ups. Interesting stuff!
> correct and uncontraversial
From the original quote it is clear he was referring to the idea of aliens being detectable by infrared because they will absorb all of their sun's energy. Later in the same paragraph he says:
> Unfortunately I went on to speculate about possible ways of building a shell, for example by using the mass of Jupiter... > These remarks about building a shell were only order-of-magnitude estimates, but were misunderstood by journalists and science-fiction writers as describing real objects. The essential idea of an advanced civilization emitting infrared radiation was already published by Olaf Stapledon in his science fiction novel Star Maker in 1937.
So the Dyson Sphere is a rhetorical vehicle to make an order-of-magnitude estimate, not a description of a thing that he thought could physically exist.
Full quote from the video cited before "the idea was a good one":
> science fiction writers got hold of this phrase and imagined it then to be a spherical rigid object. And the aliens would be living on some kind of artificial shell. a rigid structure surrounding a star. which wasn't exactly what I had in mind, but then in any case, that's become then a favorite object of science fiction writers. They call it the Dyson sphere, which was a name I don't altogether approve of, but anyway, I mean that's I'm stuck with it. But the idea was a good one.
Again he explicitly says this "wasn't exactly what I had in mind." This one hedges a bit more and could be interpreted as his saying the idea of a Dyson Sphere is a good one. He may have meant that in the sense of it being a good science fiction idea though, and he subsequently goes on to talk about that.
The Dyson Sphere is good for order-of-magnitude calculations about hypothetical aliens, and also for selling vapourware to the types of people who uncritically think that vapourware is real.
Have you read the paper itself, not just summaries of the idea? It's obvious from the way he wrote it, dripping in sarcasm. Talking about "Malthusian principles" and "Lebensraum", while hand waving away any common sense questions about how the mass of Jupiter would even be smeared into a sphere around the sun, just saying that he can conceive of it and therefore we should spend public money looking for it. He's having a lark.
Also, he literally said it was a joke, and was miffed that he was best know for something he didn't take seriously.
https://www.youtube.com/watch?v=fLzEX1TPBFM
Yeah, that's the point ... it's stupid to believe humanity is capable of deploying that much infrastructure. We cannot do even 0.01% of it.
What do you think the limiting factor is? I don't see why we can't scale manufacturing of satellites up as far as we want. If we mine out a substantial fraction of the mass of the earth, we can go harvest asteroids or something.
>> Dyson Sphere
> What do you think the limiting factor is?
You need to be able to harness enough raw material and energy to build something that can surround the sun. That does not exist in the solar system and we do not yet have the means to travel further out to collect, move, and construct such an incredibly huge structure. It seems like a fantasy.
The inner planets contain enough mass to create a shell of 1 AU radius with mass of 42 kg/m^2. That sounds like a plausible thickness and density for a sandwich of photovoltaics - GPUs - heat sinks.
You don't build a rigid shell of course, you build a swarm of free-floating satellites in a range of orbits.
See https://www.aleph.se/Nada/dysonFAQ.html#ENOUGH for numbers.
I am dying to know where you’ll get the energy and manufacturing scale in order to achieve this with current, or current+50-years technology.
Do tell.
The energy to build the system comes from the partial assembled system, plus some initial bootstrap energy. It grows exponentially. We seem to have enough today to build small factories in orbit.
The manufacturing scale comes from designing factory factories. They aren't that far in the future. Most factory machinery is made in factories which could be entirely automated, so you just need some robots to install machines into factories.
I was told ca. 2003 or so that because features on computer chips were getting smaller at some rate, and processor speed was getting faster at some other rate, that given exponential this or that I'd have tiny artificial haemo-goblins[1] bombing around my circulatory system that would make me swim like a fish under the sea for hours on end. But it turned out to be utter bullshit. Just like this.
[1] https://www.writingsbyraykurzweil.com/respirocytes
Great. Now run the numbers to find the energy required to disassemble the planets and accelerating the pieces to their desired locations. For reference, it takes over 10 times of propellant and oxidant mass to put something in LEO.
The burned propellant and oxygen mass (as H2O and CO2) almost all ends up back in the atmosphere when you launch to LEO, so you can keep running electrolysis (powered by solar) to convert it back to fuel.
Sure, but if we're talking about solar engineering, that mass is going to be dispersed in orbit around the sun. You're not going to be reaccumulating that any time soon.
Also it's gravitationally unstable, like Dyson Rings, where as soon as you have any perturbance from the center means that the closer side is more attracted to the sun so it enters a feedback loop.
There are only so many people who can make satellites; there are only so many things to make satellites out of; and there are only so many orbits to put them in. There are only so many reasons why a person might want a satellite. There are only so many ways of placing satellites in orbit and each requires some amount of energy, and we have access to a finite amount of energy over time.
Finally, if we limited ourselves to earth-based raw materials, we would eventually reach a point where the remaining mass of the earth would have less gravitational effect on the satellite fleet than the fleet itself, which would have deleterious effects on the satellite fleet.
Seven reasons are intuitive; I’m sure there are many others.
People can build a factory that makes satellites. And then a factory that makes factories to make satellites.
There is plenty of material in the solar system (see my other response), and plenty of orbits, and launch capability can scale with energy harvested so the launch rate can grow exponentially.
Lots of people will probably decide they don't want any more satellites. But it only takes a few highly determined people to get it done anyway.
>Just imbest[1] and it will grow exponentially.
That's how that argument sounds like, particularly when you hear it from someone who is as broke as it can be.
It's easy to type those ideas in a comment, or a novel, or a scientific paper ... bring them to reality, oh surprise! that's the hard part.
1: The dumb version to invest
> Finally, if we limited ourselves to earth-based raw materials, we would eventually reach a point where the remaining mass of the earth would have less gravitational effect on the satellite fleet than the fleet itself, which would have deleterious effects on the satellite fleet.
The Earth's crust has an average thickness of about 15-20 km. Practically we can only get at maybe the top 1-2 km, as drill bits start to fail the deeper you go.
The Earth's radius is 6,371 km.
So even if we could somehow dug up entire crust we can get to and flung it into orbit, that would barely be noticeable to anything in orbit.
Once you dig up the top kilometer of a planet's crust, what's under your feet? The next kilometer!
That would suck to do to Earth, but we can launch all of Mars's mass into the swarm.
After a few decades, you need to start replacing all the solar panels.
And the robot army being used to do the construction and resource extraction will likely have a much shorter lifespan. So needs to be self-replicating/repairing/recycling.
The physical amount of material in the solar system is a pretty big limiting factor.
Yeah, but besides not having the physical amount of material available in the solar system, or the availability of any technology to transfer power generated to a destination where it can serve a meaningful purpose in the foreseeable future, or having the political climate or capital necessary for even initiating such an effort, or not being able to do so without severely kneecapping the habitability of our planet, there are aren't really any meaningful barriers that I can see.
Are you suggesting that beggars would ride, if only wishes were horses!?
In 2026? Grift.
But the factory ~~can~~must grow.
We will have cyber taxis and FSD 100% next year.
Doesn't this risk some unforeseen effects on Earth or the rest of the solar system at that scale? Disruption of magnetic shield, some not yet known law of physics suddenly getting felt etc.?
It's not really going to happen so we don't have to worry about that.
Photovoltaic production has been doubling every year. That's not a huge amount of doubling!
Pfft that would just require setting up an entire lunar mineral extraction and refining system larger than we have on earth, just minor details.
Yeah
I don't know where this delusion of "Servers in space" came from, I think of it as the new NFTs
But I bet those pushing for it are very interested in feeding the grift
Help me understand something. We make 1 TW of cells per year but we're struggling with bringing 1 GW consuming data centers online?
Nameplate capacity needs a derate for availability, so you can drop it down to about 200GW(e) equivalent continuous power assuming we're making and deploying enough batteries to support it. More, obviously, if those panels are going to an equatorial desert, less if they're going to sunny Svalbard in the winter time.
A former NASA engineer with a PhD in space electronics who later worked at Google for 10 years wrote an article about why datacenters in space are very technically challenging:
https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
I don't have any specialized knowledge of the physics but I saw an article suggesting the real reason for the push to build them in space is to hedge against political pushback preventing construction on Earth.
I can't find the original article but here is one about datacenter pushback:
https://www.bloomberg.com/opinion/articles/2025-08-20/ai-and...
But even if political pushback on Earth is the real reason, it still seems datacenters in space are extremely technically challenging/impossible to build.
The real reason is, Elon has SpaceX and xAI. He can create an illusion of synergy and orders of magnitude advancements to boost the market cap and pocket all the money. He realized long time ago you don't need to deliver to play the market cap game, in fact it's better if you are selling a story far in the future rather than a something you can deliver now.
exactly. This smells like a way to boost the SpaceX IPO to meme-stock premiums
We don’t even have a habitable structure in space when the ISS falls, there is no world in which space datacenters are a thing in the next 10, I’d argue even 30 years. People really need to ground themselves in reality.
Edit: okay Tiangong - but that is not a data center.
Who is “we”? https://en.wikipedia.org/wiki/Tiangong_space_station
Good point. Still a long, long way from data centers.
I don't think any of the companies that say they are working on space data centers intend them to be habitable.
> We don’t even have a habitable structure in space
Silicon is way more forgiving than biology. This isn’t an argument for this proposal. But there is no technical connection between humans in space and data centers other than launch-cost synergies.
Okay, but a human being represents what, 200 W of power? The ISS has a crew of 3, so that's less than a beefy single user AI workstation at full tilt. If the question is whether it's practical to put 1-2 kW worth of computing power in orbit, the answer is obviously yes, but somehow I don't think that's what's meant by "datacenter in space".
I don't know, 10 years seems reasonable for development. There's not that much new technology that needs to be developed. Cooling and communications would just require minor changes to existing designs. Other systems may be able to be lifted wholesale with minimal integration. I think if there were obstacles to building data centers on the ground then we might see them in orbit within the next ten years.
I don't see those obstacles appearing though.
The same things you are saying about data centers in space was said by similar people 10-15 years ago when Elon musk said SpaceX would have a man on Mars in 10-15 years.
We have had the tech to do it since the 90's, we just needed to invest into it.
Same thing with Elon Musks hyperloop, aka the atmospheric train (or vactrain) which has been an idea since 1799! And how far has Elon Musks boring company come to building even a test loop?
Yeah, in theory you could build a data center in space. But unless you have a background in the limitations of space engineering/design brings, you don't truly understand what you are saying. A single AI data center server rack takes up the same energy load of 0.3 to 1 international space station. So by saying Elon musk can reasonable achieve this, is wild to anyone who has done any engineering work with space based tech. Every solar panel generates heat, the racks generate heat, the data communication system generates, heat... Every kW of power generated and every kW of power consumes needs a radiator. And it's not like water cooling, you are trying to radiate heat off into a vacuum. That is a technical challenge and size, the amount of tons to orbit needed to do this... Let alone outside of low earth... Its a moonshot project for sure. And like I said above, Elon musk hasnt really followed through with any of his moonshots.
> A single AI data center server rack takes up the same energy load of 0.3 to 1 international space station.
The ISS is powered by eight Solar Array Wings. Each wing weighs about 1,050kg. The station also has two radiator wings with three radiator orbital replacement units weighing about 1,100kg each. That's about 15,000 kg total so if the ISS can power three racks, that's 5,000kg of payload per rack not including the rack or any other support structure, shielding, heat distribution like heat pipes, and so on.
Assuming a Falcon Heavy with 60,000 kg payload, that's 12 racks launched for about $100 million. That's basically tripling or quadrupling (at least) the cost of each rack, assuming that's the only extra cost and there's zero maintenance.
Falcon Heavy does not cost 100M when launching 60 metric tons.
At 60 metric tons, you're expending all cores and only getting to LEO. These probably shouldn't be in LEO because they don't need to be and you probably don't want to be expending cores for these launches if you care about cost.
The real problem typically isn't weight, it's volume. Can you fit all of that in that fairing? It's onli 13m long by 5m diameter...
His time estimates are notoriously, um, aggressive. But I think that's part of how his companies are able to accomplish so much. And they do, even if you're upset they haven't put a human on Mars fast enough or built one of his side quests.
"We specialize in making the impossible merely late"
> Cooling and communications would just require minor changes to existing designs.
"Minor" cooling changes, for a radically different operating environment that does not even have a temperature, is a perfect insulator for conduction and convection, and will actively heat things up via incoming radiation? "Minor" ? Citation very much lacking.
We also don't have fully reusable launch vehicles, yet. But we will shortly. That will decrease the cost of launch by at least an order of magnitude.
Still there will be a lot of engineering problems to solve.
2-3 years seems very short, but 10 years seems long to me.
Ok then short SpaceX stock when it IPOs.
^ the Left don’t even hide their malevolence anymore
“Markets can remain irrational longer than you can remain solvent.” - John Maynard Keynes
What does stock price have to do with anything?
That someone could put a data center in space for the price of 100 years of eliminating world hunger doesn’t mean shit.
People always make this claim about world hunger elimination with no sources. Keep in mind we make more than enough calories to feed everyone on the planet many times over, it's a problem of distribution, of getting the food to the right areas and continuing cultivation for self sufficiency.
That’s right, it’s an allocation of resources problem, and some people seem to control almost all the resources.
Even the most magnanimous allocators cannot defeat the realities of boots on the ground in terms of distribution. It is a very difficult problem that cannot be solved top down, the only solution we've seen is growth of economic activity via capitalistic means, lifting millions, billions out of poverty as Asia has done in the last century for example.
I argue that if you have literal hundreds of billions of hard cash to burn for stupid things like AI datacenters, you could afford to make the lives of millions of starving people not suck instead, pretty easily so. But to do that, you'd have to try, and that would mean actually doing something good for humanity. Can't have that as a billionaire.
Ok but what if I shoot a car into space and buy my own social media company. Surely thats a better use of billions!
Why would you short the stock?
If you're hellbent on arguing with a cult, it will be much cheaper to go down to your local Church of Scientology and try to convince them that their e-meter doesn't work.
As if company performance actually affected stock price when it comes to anything Elon Musk touches.
For fuck's sake, TSLA has a P/E of a whopping *392*. There is zero justification for how overvalued that stock is. In a sane world, I should be able to short it and 10x my money, but people are buying into Musk's hype on FSD, Robotaxi, and whatever the hell robot they're making. Even if you expected them to be successes, they'd need to 20x the company's entire revenue to justify the current market cap.
It's much easier to find a country or jurisdiction that doesn't care about a bunch of data centers vs launching them into space.
I don't get why we aren't building mixed use buildings, maybe the first floor can be retail and restaurants, the next two floors can be data centers, and then above that apartments.
In Switzerland infomaniak built a data center under apartments and DC heat is used for heating. There are some videos about it.
Americans have trouble understanding something like that. We believe anything short of a 3bdrm house with a lawn and backyard is communism.
I'd love to live in a dense city. My office within waking distance. A Cafe in my apartment building, etc.
The US has district heating systems. The country is very big and varied, as much as people like to paint it as homogenous.
> I'd love to live in a dense city. My office within waking distance. A Cafe in my apartment building, etc.
Then move to one?
Probably for the same reasons they aren't doing mixed use prison and restaurant buildings.
What you don't want to live near the newest poisonous abomination that the whiz kids dreamt up? Do you want China to take over America or something?
Data centers don't do anything other than sit there and turn electricity into heat. They only emit nothing but heat (which could be useful to others in the building).
In America they have "temporary" jet turbines parked next to them burning gas inefficiently with limited oversight on pollution and noise because they are "temporary".
Mixed-use buildings with restaurants on the lower floors and residential on the upper floors are very common. Not sure what prisons have to do with anything.
It's not just "very challenging", it's "very challenging and also solves no actual problem we face".
> A former NASA engineer with a PhD in space electronics who later worked at Google for 10 years wrote an article about why datacenters in space are very technically challenging
It's curious that we live in a world in which I think the majority of people somehow think this ISN'T complicated.
Like, have we long since reached the point where technology is suitably advanced to average people that it seems like magic, where people can almost literally propose companies that just "conjure magic" and the average person thinks that's reasonable?
It's just the thought process that comes with shallow understanding:
you can’t tell me the microwave isn’t magic. it’s magic.
I can put things in a box that uses spooky electromagnetic waves to tickle water molecules to the point that they get hot and maybe boil off, given the chance? Sounds like magic to me
Google is currently working on AI data centers in space.
https://blog.google/innovation-and-ai/technology/research/go...
Nice article, the first one. I hope they try it, burn many billions of cash, and then fail. I also hope they don't spread radioactive material across the whole atmosphere when failing, though.
It's not like launching stuff into space doesn't have pushback, either. See: starlink satellites.
"Technically challenging", a nice way to say "impossible"
Just like rockets landing themselves
No, rockets landing themselves is just controlling the mechanism you use to have them take off, and builds on trust vectoring technology from 1970s jet fighters based on sound physics.
Figuring out how to radiate a lot of waste heat into a vacuum is fighting physics. Ordinarily we use a void on earth as a very effective _insulator_ to keep our hot drinks hot.
Figuring out how to radiate a lot of waste heat into a vacuum is just building very large radiators.
> Figuring out how to radiate a lot of waste heat into a vacuum is fighting physics.
Radiators should work pretty well, and large solar panels can do double duty as radiators.
Also, curiously, newer GPUs are developed to require significantly less cooling than previous generations. Perhaps not so coincidentally?
Well there lies the rub, solar panels already need their own thermal radiators when used in space ...
Great, so you seem to agree the technology exists for this and it is a matter of deploying more of it?
It's a matter of deploying it for cheaper or with fewer downsides than what can be done on earth. Launching things to space is expensive even with reusable rockets, and a single server blade would need a lot of accompanying tech to power it, cool it, and connect to other satellites and earth.
Right now only upsides an expensive satellite acting as a server node would be physical security and avoiding various local environmental laws and effects
> Right now only upsides ...
You are missing some pretty important upsides.
Lower latency is a major one. And not having to buy land and water to power/cool it. Both are fairly limited as far as resources go, and gets exponentially expensive with competition.
The major downside is, of course, cost. In my opinion, this has never really stopped humans from building and scaling up things until the economies of scale work out.
> connect to other satellites and earth
If only there was a large number of satellites in low earth orbit and a company with expertise building these ;)
I mostly agree with you, but I don't understand the latency argument. Latency to where?
These satellites will be in a sun-synchronous orbit, so only close to any given location on Earth for a fraction of the day.
> And not having to buy land and water to power/cool it.
It's interesting that you bring that up as a benfit. If waterless cooling (i.e. closed cooling system) works in space, wouldn't it work even better on Earth?
You need to understand more of basic physics and thermodynamics. Fighting thermodynamics is a losing race by every measure of what we understand of the physical world.
> Fighting thermodynamics is a losing race
The great thing about your argument is that it can be used in any circumstance!
Cooling car batteries, nope can't possibly work! Thermodynamics!
Refrigerator, are you crazy? You're fighting thermodynamics!
Heat pump! Haah thermodynamics got you.
Actually all of those things agree with the same laws that dictate why data centers can't work in space.
Your examples prove our case. You just must not understand how they work
1kW TDP chips need LESS cooling?
Yes, Rubin reportedly can deal with running significantly hotter.
That makes radiating a much more practical approach to cooling it.
I see what you’re saying - higher design temp radiates better despite more energy overall to dissipate.
> I see what you’re saying - higher design temp radiates better despite more energy overall to dissipate.
Yes, running hotter will cause more energy to be radiated.
but
These parts are not at all designed to radiate heat - just look at the surface area of the package with respect to the amount of power they consume.
This is a classic case of listing all the problems but none of the benefits. If you had horses and someone told you they had a Tesla, you'd be complaining that a Tesla requires you to dig minerals where a horse can just be born!
His point is that everyone said landing and reusing rockets was impossible and made fun of Elon and SpaceX for years for attempting it.
The difference is that it was mostly clueless people like Thunderf00t who said it was impossible, who nobody took seriously. I don’t remember that basically all relevant experts claimed it was near impossible with current technology. That’s the situation now.
There’s also fairly clear distinction with how insane Elons plan has become since the first plans he laid for Tesla and SpaceX and the plans he has now. He has clearly become a megalomaniac.
Funnily enough, some of the things people said about Tesla is coming true, because Elon simply got bored of making cars. It’s now plausible that Tesla may die as a car company which I would not have imagined a few years ago. They’re arguably not even winning the self driving and robotics race.
> landing and reusing rockets
Currently SpaceX have managed to land the booster only, not the rocket itself, if you are thinking about Starship. And reusability of said rocket is also missing (collecting blown up pieces from the bottom of the ocean doesn't count!).
No, people made fun of Elon for years because he kept attempting it unsafely, skirting regulations and rules, and failing repeatedly in very public ways.
The idea itself was proven by NASA with the DC-X but the project was canceled due to funding. Now instead of having NASA run it we SpaceX pay more than we'd ever have paid NASA for the same thing.
DC-X test flight: https://www.youtube.com/watch?v=gE7XJ5HYQW4
It's awesome that Falcon 9 exists and it is great technology but this guy really isn't the one anyone should want in charge of it.
>Now instead of having NASA run it we SpaceX pay more than we'd ever have paid NASA for the same thing.
This doesn’t pass the smell test given that the cost of launch with spacex is lower than it ever was under ULA.
NASA has never been about cheap launches, just novel technology. Look at the costs of Saturn and SLS to see what happens when they do launch.
He also said he could save the us a trillion dollars per year with DOGE, and basically just caused a lot data exfiltration and killed hundreds of thousands of people, without saving any money at all
Elon Musk killed hundreds of thousands of people?
Yes. Mostly kids, because of the DOGE ransacking of USAID
https://healthpolicy-watch.news/the-human-cost-one-year-afte...
He said impossible, this was done recently, by spacex themselves.
Yeah but he’s an expert his opinion can be dismissed bro this is 2026
> It(Solar) works, but it isn't somehow magically better than installing solar panels on the ground
Umm, if this is the point, I don't know whether to take rest of author's arguments seriously. Solar only works certain time of the day and certain period of year on land.
Also there is so limited calculations for the numbers in the article, while the article throws of numbers left and right.
> Solar only works certain time of the day and certain period of year on land
The same goes for LEO!
Most space datacenter plan plans to use sun-synchronous orbit.
No one is interested in excuses on why it can't be done. Were in interested in the plan on how they plan to do it.
The guy is saying satellite communication is restricted to 1Gbps ffs. SpaceX is way past that.
SpaceX is too big to fail. It's important for national security.
I wonder if Elon wants to tangle all his businesses into SpaceX so they are all kept afloat by SpaceX's importance.
Elon can't legally financially entangle Tesla to SpaceX due to Tesla being a public company, so his hands are tied.
Tesla is clearly benefiting from protectionism and its sales would collapse if BYD were allowed to openly sell in the US. Most people just want affordable, maintainable and reliable cars.
> Elon can't legally financially entangle Tesla to SpaceX due to Tesla being a public company, so his hands are tied.
He absolutely could do it, just like he did when Tesla bought SolarCity. It just isn’t as easy when one of the companies is public than when both are private.
Tesla to invest $2B in Elon Musk’s xAI https://techcrunch.com/2026/01/28/tesla-invested-2b-in-elon-...
We're witnessing a bailout and downloading of costs, at scale. Whether or not one buys into whatever the vision of these companies are - it's clear, there's interdealing.
Tesla theoretically now owns a chunk of xAI... whose valuation will no doubt increase due to the internalized SpaceX acquisition. Append to this a future IPO, as discussed in the artice, presumably an eventual premium of 20-50% (reasonable, 14% purely for the ibankers when this will happen)... yields to an interesting bailout situation.
To me, the real question is why. The $2B from Tesla can't possibly move the needle for any party involved in this transaction. If this were to be work 50x as opposed to a potential 50% upside (hell, make it 2x for argument's sake) it still doesn't compute. So what's the actual reason.
I don't know, but it could be long term vs short term capital injection.
> Tesla is clearly benefiting from protectionism and its sales would collapse if BYD were allowed to openly sell in the US
So would most of EU car makers in Europe. China is not playing by the same rules and everyone with car manufacturing domestically is slamming them with tariffs.
How isn't China playing by the same rules? Every country subsidises and supports industry it thinks is important, surely nothing would stop Germany from investing into Volkswagen and BMW or the US from investing into Ford the same way China invests into BYD?
I’m old enough to remember when this was said about Solar City
Got that double digit age locked down I see, congrats!
> Elon can't legally financially entangle Tesla to SpaceX
Bill Ackman has proposed taking SpaceX public by merging it with his Pershing Square SPARC Holdings, distributing 0.5 Special Purpose Acquisition Rights (SPARs) to Tesla shareholders for each share held. Each SPAR would be exercisable for two shares of SpaceX, aimed at enabling a 100% common stock capitalization without traditional underwriting fees or dilutive warrants.
With SpaceX IPO set to be one of the biggest of all time, this could have a pretty gnarly financial engineering impact on both companies -- especially if the short interest (direct or through derivatives) remains large.
Why would SpaceX go public? They already have a robust enough private market to give liquidity to all of their employees and shareholders who want it. They can get more private investment.
Going public would add a lot of hassle for little to no gain (and probably a negative of having to reveal their finances).
It has been widely reported for weeks that SpaceX is planning to go public in a few months. The reason is they have big plans to run a vast network of AI servers in orbit and will need to raise a massive amount of funding. xAI merger fits with that plan. I'd assume SpaceX still plans to go public.
Was ignored on HN but here's an article explaining:
https://arstechnica.com/space/2025/12/after-years-of-resisti...
> a vast network of AI servers in orbit
That story makes no technical sense. There's no benefit to doing this. Nobody should believe it any more than boots on Mars by 2030.
Or any more than "full self driving" by 2017.
it wasn't ignored on HN, there were many articles correctly noting that building data centers in space is a stupid stupid idea because cooling things there is infeasible
Was doing some back of the envelope math with chatGPT so take it with a grain of salt, but it sounds like in ideal conditions a radiator of 1m square could dissipate 300w. If this is the case, then it seems like you could approach a viable solution if putting stuff in space was free. What i can't figure out is how the cost of launch makes sense and what the benefit over building it on the ground could be
What temperature were you assuming?
Because the amount of energy radiated varies with the temperature to the fourth power (P=εσT^4).
Assuming very good emissivity (ε=0.95) and ~75C (~350K) operating temperature I get 808 W/m2.
They would most likely launch with TPUs designed for space and target lower temperatures, closer to 60C.
Google, Blue Origin and at least 5 other smaller companies have announced plans to build data centers in space. My understanding is the cooling issue is not the show stopper you assume.
yup, bezos said "we will be able to beat the cost of terrestrial data centers in space in the next couple of decades". presumably this means they'll need huge ass radiators, so its all about bringing down launch costs since they'll need to increase mass.
lol WHAT?
AI datacenters are bottlenecked by power, bandwidth, cooling, and maintenance. Ok sure maybe the Sun provides ample power, but if you are in LEO, you still have to deal with Earth's shadow, which means batteries, which means weight. Bandwidth you have via starlink, fine. But cooling in space is not trivial. And maintenance is out, unless they are also planning some kooky docking astromech satellite repair robot ecosystem.
Maybe the Olney's lesions are starting to take their toll.
Weirdest freaking timeline.
The shadow thing can be solved by using a sun-synchronous orbit. See for example the TRACE solar observation satellite, which used a dawn/dusk orbit to maintain a constant view of the sun.
Cooling, on the other hand? No way in hell.
Every telco satellite can cool its electronics. However, more than a few kW is difficult. The ISS has around 100kW and is huge and in a shadow half the time.
> Cooling, on the other hand? No way in hell.
Space is actually really cold when the sun is blocked
So, solar panels on side, GPUs on the other, maybe with a big ass radiator ...
Space is empty, not cold.
The cooling is the bit where I'm lost on, but it will be interesting to see what they pull off. It feels like everyone forgets Elon hires very smart people to work on these problems, it's not all figured out by Elon Musk solely.
Google, Blue Origin and a bunch of other companies have announced plans for data centers in space. I don't think cooling is the showstopper some assume.
Good call out, and really interesting. SpaceX being the cheapest way to get things into space, it seems like SpaceX is about to become extremely lucrative.
He's broken pretty much all the other financial rules.... for example, the amount of blatant self-dealing he gets away with is staggering.
As long as the consequences of his actions continue to increase the paper value for investors, regulations don't really have teeth because there aren't damages. So the snowball gets bigger and the process repeats.
I've been thinking about this recently as I hear it often. Would people who want to buy a car in the Tesla price range really choose a slightly cheaper Chinese EV if those were available?
Personally I have a hard time believing this. But even if you had similarly priced Chinese options, I would guess the main reason for buying a Tesla is not just because you want an EV. While a Tesla will be a reliable baseline EV, surely the reason you (or at least I) would buy one is for the supervised self-driving feature.
BYD are just affordable and maybe reliable, regarding maintenance their spares are hard to come by and are almost as hard to work with as Tesla and other brands.
I've done plenty of work on my own Tesla. It's not hard to work on at all. Parts are not even very difficult. There are plenty of 3rd party shops (such as one I went to when I needed to replace my windshield.) I really wonder why people continue to think this. It's not 2016 any more.
Tesla body work is extremely expensive. Aluminum, extensive welding instead of fasteners, substantially reduced modularity due to castings, specialized tooling just off the top of my mind.
Are you a car mechanic living in China?
Presumably "hard to come by" would be somewhat irrelevant in any jurisdiction other than the US?
Did you see how this last quarter where BYD sales fell off a cliff?
[Nearly] all is possible when you have a board of simps/cultists
It's "ironic?" considering Tesla launching in China is what created the necessary supply chain to turn BYD into the powerhouse it is today. Tesla's greed will become their own demise.
Tesla cars made in Shanghai are sold in Europe and other places. That is helping them be competitive and they haven't had much price pressure until recently. Just because the Chinese have their own internal competition and deflation which drove their prices down aggressively doesn't mean it was a bad idea to build there. Also the idea the Chinese couldn't figure it out without an American company coming there first to show them is pretty silly.
Tesla Shanghai opened in 2019
BYD made their first hybrid in 2008 and they were a battery company since the 90s
Starship has a large number of critical milestones coming: Can it land and quickly reuse the upper stage? If not, it can't make refueling flights without building a dozen or two starships. Can it carry the full specified payload? If not, it can't even try to refuel in orbit. If it can't refuel in orbit, it can't go beyond earth orbit. Etc.
Everything has to go right or it will be irrelevant before it works.
> Everything has to go right or it will be irrelevant before it works.
Starship is not all of SpaceX. Saying, maybe because one hates Musk, that SpaceX is going to become irrelevant is wishful thinking.
In 2025 SpaceX launched more rockets into space than the entire world ever sent in a year up to 2022, something crazy like that.
Then out of, what, 14 000 active satellites in space more than half have been launched by SpaceX.
SpaceX is, so far, the biggest space success story of the history of the human race (and GP is right in saying that SpaceX is now a national security matter for the US).
Model S was the most successful EV. If you think cybercab is the vehicle of the future, look at the timeline of the only robo taxi in commerce in the US.
Everything has to go right with that, or cybercab will be irrelevant before it works. Same deal. Same bullshitter.
Model S was successful until Model 3/Y blew it out of the water. Waymo’s timeline is not relevant because they lose money on every car and every deployment. Tesla’s the only financially successful developer of self driving. They can scale it up much faster.In fact, instead of making $5k per car produced, cybercab will net them $50k per car per year.
The world doesn't consist of just Waymo and Tesla, and even if it did there's no guarantee either succeeds.
> cybercab will net them $50k per car per year.
Assuming no mass boycotts, nor targeted vandalism. We've already seen both in the last 12 months.
It’ll be fine. Especially when people compare the price of ownership/uber to robotaxis.
When people actually compare prices, they note that Chinese cars also have autopilot and cost less than half of a Tesla, new.
What's keeping Chinese brands out of the USA, isn't keeping them out of Europe or much of anywhere else.
Tesla remains competitive in China, which can't be said of European EVs. Chinese ADAS are much better than European ones but still far behind FSD.
To bring the discussion back on topic: $50k/year or ~$250k over the course of the vehicle's lifetime, instead of $5k for a singular sale event, is why the path for the company is crystal clear. Cybercab is the same kind of step for Tesla as the Model 3 was back in 2017.
More likely that it's going to be the same kind of step for Tesla as the Oculus was for Facebook.
I’ll grant you that it could be, and I’m betting it won’t while you are betting it will. The future is now obvious to fsd14 and robotaxi users. Failure is no longer likely.
Yeah and it’s going to bankrupt VW/Stellantis. Surprised Europeans just don’t seem to give a damn about that.
BYD sales in January 2026 are down 30% YoY. Not looking great for them in 2026.
> Tesla’s the only financially successful developer of self driving.
This is completely false. Audi and Chevrolet both have self driving as good as Tesla.
This would literally only be said by a person who hasn’t tried fsd14 so do yourself a favor and go for a test ride of a Tesla and Audi/Chevrolet and report back. They are not comparable.
They sure as hell do not. SuperCruise only worked in pre mapped areas and bails whenever there's construction or deviation to plan. It's analogous to Tesla AP2 at best.
FSD works EVERYWHERE, almost any time.
97% of their sales are model 3 / Y
> In 2025 SpaceX launched more rockets into space than the entire world ever sent in a year up to 2022, something crazy like that.
Not just that, the cost of each rocket launch is drastically cheaper than all of its competitors costs.
How vital is it really to national security? Starlink will have competition from Amazon Leo in the next few months. And while SpaceX is obviously in the lead in launch capability with Starship, there are multiple launch providers capable of providing roughly the same services the Falcon 9 and Heavy provide today.
The same services as Falcon 9 are 20x the cost and launch 1/20th as much as well. That's like producing hand made good in America versus via a manufacturing line in China.
> Starlink will have competition from Amazon Leo in the next few months
Amazon Leo will have 14k satellites in space in a few months? Wow! Amazing!
Why is it too big to fail? SpaceX can be dissected, parts be sold to the government or the competition.
It's too big to fail for Musk, because it is one source of his money, in large paid by the US tax payer.
I think he will spin Tesla off since electrification and autonomy are no longer cool (he can’t build good quality cars or reliable FSD)
Haven't you heard? Tesla is pivoting to building humanoid robots instead. They haven't sold a single one, but it toootally warrants retooling their car factories, pinky promise!
FSD is incredibly reliable. Build quality of US built cars is middle of the pack, Europe/China built Teslas are top of the pack.
Incredible shilling, bravo
Oh c'mon now. Damn model 3 and model S I have driven were considerably lower quality interiors than an ass end Citroen or Fiat. The Model S, a 2023 model the doors didn't even fit properly. And that was all Europe.
As for FSD, nope. Unless you redefine the word reliable.
Edit: I owned a 2018 Model S as well. Literally the worst fucking car I have ever owned or driven.
I disagree. Model 3 has soft touch everywhere. Freaking bmw 3 series has plastic on most frequently touched bits.
Since you are in europe you have no idea how good fsd is.
BMW actually has a reasonable control surface though, not a grand user interface experiment by some crack heads.
As I'm in Europe I just get trains.
The bmw interface is the actual fucking joke. Everything you need on Teslas is accessible from the steering wheel in addition to the touchscreen.
Apart from the speedometer which is outside your safe FOV in the Tesla.
And everything in the BMW you should be dealing with when driving is on or around the steering wheel.
That’s not how human eyes work. https://en.wikipedia.org/wiki/Vision_span
> That’s not how human eyes work. https://en.wikipedia.org/wiki/Vision_span
The article you linked agrees with me. Greatest resolution in the macula which is a span of approximately 6 degrees from the centre.
Sigh...
No, it agrees with me. You’re NEVER looking at the road and reading your speed at the same time.
Unless you have a modern car with a HUD
Unless it’s obstructing the view of the road, not really
> BMW actually has a reasonable control surface though, not a grand user interface experiment by some crack heads.
Really? It's one thing to hate Elon Musk, but you're talking about a lot of brilliant engineers who worked on these cars, everything from the components to the software. It's uneeded low blow just because you don't like Elon Musk.
The UX is a mess. Why does the car always label the trunk as open rather than have a button that I press to open it?
Why does cruise control sometimes change to the speed limit and sometimes not?
Why does auto lane change sometimes need me to start the manoeuvre and sometimes not? If I guess wrong and start the lane change myself, all autopilot just disengages suddenly.
I have to proove that I'm holding the wheel by wiggling it from time to time, but if I accidentally wiggle too hard it disengages. Why not have a sensor or use the cameras to detect if I'm holding the wheel?
My son didn't shut the back door properly. I started driving and the car started binging. It didn't tell me why it was binging until I put it in park and looked at the pretty 3d representation of the car, then noticed that the door was open.
Maybe if I drove more regularly I would get used to all this stuff. The car was borrowed and I gave it back.
I’m glad you found a place to get these complaints off your chest, but these are kind of hilarious. the button says “open trunk”. It’s a verb. If this is your complaint then lmao have you not seen what other OEM software looks like? Door open doesn’t just ding, it shows a warning with plain english explanation and an icon.
For the rest of your complaints you can mostly thank the overzealous EU/unece regulation which limits steering torque and requires intervention. FSD has none of those concerns, it just drives and does not require torque on the wheel.
Looking at cyber truck I can’t help but disagree with you. Absolutely questionable design choices. From top to bottom.
So the Cybertruk is one vehicle out of an entire line up, I get not liking one model but what's that go to do with the entire line up?
That is what they want you to think it isn't too big to fail there are plenty of competitors with much stronger engineers and principles than this grifter.
Of course it isn't "too big to fail". Even banks aren't. Despite recent history large banks have failed often throughout history. There's no such thing. It may take down the supporting sovereign government (Dutch East Indies) but life goes on and new political orgs appear. People be people.
Too big to fail is a very recent modern myth. Go back 100+ years and lots of banks failed leading into the Great Depression.
Every system has a break point.
Right. You do have a point, and I think Dutch East Indies is a good example, but I feel this is discussing semantics. Too big to fail, I interpret in this situation as the government having a strategic reason to keep it afloat so it will probably prop it up in case something goes wrong. This makes it have a much more stable position.
Simply put, yes.
This is not good for SpaceX. It's a less valuable company with X and xAI. But it helps Elon make it look like he runs two successful businesses.
Let’s be honest - this is just a way to prop up Twitter/X. It makes SpaceX shareholders subsidize X, and also American taxpayers who are giving contracts to SpaceX for highly sensitive things. The government should ideally refuse to give SpaceX work unless it unwinds this.
Why? The government is paying less for SpaceX than alternatives. It th cheapest and best service.
Because Twitter/X is distorting our politics (with ann unbalanced scheme of censorship / amplification / suppression) and destroying the country by mainstreaming far right supremacist politics. Twitter/X does not deserve a single dollar of taxpayer money. If SpaceX is now part of that machine, it doesn’t deserve a single dollar either. I would rather pay more for alternatives and encourage their growth. I also look at any money given to this company as the equivalent of GOP campaign funding, so I feel it should be treated as illegal under the law.
Shouldn't the government be aiming to pay the lowest price for the best goods and services rather than using procurement as a way to promote or suppress certain political opinions?
The government is prevented from doing that by a little thing called the first amendment. "Mainstreaming far right supremacist politics" is just a hyperbolic way of saying he has politics you don't like and is exercising his freedom of the press by promoting it on the media platform he owns. Legally that is no different then the rights that every newspaper and TV station in the country has.
First of all, the current government doesn't give a shit about the first amendment and is successfully putting a chilling effect on it through various means. Both through illegally using government funding as a hammer to require independent companies to curtail their speech, or by using regulation.
Second, history will look back and realize that without taking into account the volume of your voice, you don't really have free speech in a way that matters. If you the person next to you can use a megaphone that is so loud that no one hears you, you effectively have no speech. A great many democracies implicitly realize this and thus have election spending limits tied to the number of supporters. The US, through it's lobby system, and through party affiliated control of third party networks, does not.
Musk is, indeed, allowed under the 1st to promote whatever he wants to promote. Him being a hypocrite about "free speech absolutism" is not a crime.
However, the current US administration appears to be actively violating the 1st and 5th in a bunch of ways, the 14th that one time, and making threats to wilfully violate the 2nd for people they don't like and the 22nd to get a third term. It is reasonable, not hyperbolic, to be concerned about Musk's support of this.
Actually the Trump administration is trying to strip legal status from people and deport them by way of an obscure law that gives the Secretary of State the discretion to do so if they deem those people a threat to the foreign policy goals of the US.
If these laws are still on the books when the next D administration takes over, they should use them against Elon, Thiel, etc - strip them of US citizenship, deport them, and nationalize their companies (followed with repealing those laws)
I disagree. He would be using taxpayer money to boost his preferred speech. And it is essentially campaign funding for the GOP. It should be treated as such.
I think that line of argument would work in my country of birth, the UK, but I don't think it works in the USA.
You do not lose your right to free speech by providing contractual services to the US government.
I would rather our government not get in the habit of violating the multiple laws put in place to keep it from playing favorites and picking winners.
If anything, I think this is actually the other way around - channeling crazy AI bubble money towards SpaceX, after the funding from goverment contracts has dried up. Twitter is just the icing on the cake.
Quite ingenious, you have to give Musk that. This is why he is making so much money.
It’s also a way to distract from the fact that alleged pedophile and rapist Elon had 3 underaged foreign nationals trafficked to him at the space x headquarters by convinced pedophile and rapist Jeffrey Epstein, per the Epstein files.
And our tax dollars.
SpaceX is too big to fail for sure. If it goes bankrupt, it'll be broken down, and trimmed down to the succesful launching operation. But I don't think it's the reason it's buying xAI.
SpaceX buying xAi means that xAI shareholders are cashing in on its current high valuation. It makes it look like Musk is not very confident that xAI can navigate through the AI cycle, so he might as well sell it to rake in the profits.
But he still needs control over it because of the Tesla plan and in case something else happens in the AI field that he doesn't want to miss. So he's buying it with SpaceX, because he can, freeing some of SpaceX cash to pay himself and his xAI investors.
That he managed to bullshit SpaceX investors into buying xAI is pretty crazy. But I guess that's his main talent.
SpaceX is slated to go public some time this year - June IIRC
The biggest selling point /was/ that Musk was being managed there, he wasn't tinkering with SpaceX like Twitter or Tesla, and his foolhardy direction was kept out of the company.
BUT, like Tesla, Musk cannot help himself and is making SpaceX look like a very bad investment - tying his other interests with SpaceX, allegedly using SpaceX money as a "war chest" in his battles.
There is also a danger that investors will see xAI as politically dangerous, which will really hurt SpaceX IPO
They want to go public, but have to sell the hell out of it in the meantime.
I'll bet SpaceX financials aren't as great as some people think. Remember, Elon was the guy who tried to take Tesla private, and talked a lot of smack about how silly it is to be a public company. All of a sudden he wants SpaceX to go public?
Musk has a pattern here - he used Tesla the same way, diverting resources to xAI and treating it as a funding vehicle for other ventures. Once he started doing that, Tesla's financials got murky and harder to trust. Now he's doing it with SpaceX right before the IPO. For investors, that's not 'too big to fail' protection - it's a red flag that the company finances are entangled with his personal empire instead of focused on the core business.
> The biggest selling point /was/ that Musk was being managed there, he wasn't tinkering with SpaceX like Twitter or Tesla, and his foolhardy direction was kept out of the company
The biggest selling point to who? Definitely not wall street
> SpaceX is too big to fail. It's important for national security
So was GM. Didn’t stop it from going bankrupt.
When you’re connected to Epstein, you’ll always be too big to fail
Why? Let it fail. Bring back NASA.
What do you mean "let it fail?" SpaceX has the most profitable launch system in the world and now operates >50% of all satellites in orbit. They aren't exactly in need of a bailout.
So proof of profitability is that they can shoot their own satellites into orbit?
When a company is operating at a scale where you are making orders of magnitude more orbital launches than NASA, operating a constellations of 10,000+ satellites, providing internet access to 10s of millions of people and 1 army, has raised $10s of billions in private markets at valuations in the $100s of billions, then the burden of proof is on you claiming the opposite.
The proof is that they are continuing to launch more mass into orbit than any other entity on the planet - while holding share liquidity events for their employees multiple times a year where they buy back shares. Proof is that they charge a lower cost to orbit than any of their competitors and has done so for years now.
Their revenue from Starlink is slated to be bigger than the entire NASA budget this year.
> Bring back NASA.
NASA is still here. Unfortunately, NATA fell victim to enshitification by government contracting. NASA even if it wants to simply cannot today design and launch a rocket. :(
I am no fan of Musk the man. SpaceX is a strong company and Falcon is a solid vehicle. There is not a lot of competition, and NASA trying to in-source design and supply and construction of a new, reusable LEO rocket would be a complete nightmare.
I root for a competitive rocket market, but SpaceX is at the moment critical.
You want to bring back the biggest loser? NASA kept missing deadlines for 30 years
And they still have a better track record for being on time than Elon
NASA just splurges money. The private sector is far better when it comes to money.
> The private sector is far better when it comes to money.
I've heard this a lot, but I've worked for BigCos and it seems like all they do is spend money, often superfluously. I've seen BigCos spend large quantities money on support contracts every year that haven't been used in more than a decade, or sending people on business trips across the country so they can dial into a meeting, or buying loads of equipment that sits dormant in warehouses for years and then is eventually sold off for pennies on the dollar.
I'm not convinced that they're better than the government with money allocation, I think they're just better at telling people they are.
... as we can tell by whatever the everloving fuck is going on with this press release.
I'm not talking specifically about SpaceX, although historically the cost of their rockets have been much lower than NASA. I'm being much more general. The public sector doesn't have the same incentives that private companies have, whether it's rockets or any other technology. It's sad, but it's the truth.
We have absolutely no way of gauging this until after SpaceX goes public.
SpaceX can use the same booster 30 times. NASAs new rocket can use it one time. We don't need to see financial statements to figure this one out.
I wouldn't be too sure. Depends on NASAs mission profiles and a lot of factors. Falcon heavy can bring 26.7t to GTO in expendable mode and only 8t in reusable mode. Reusable cost of Falcon is US$97 million vs US$150 million expendable.
How much does it cost to develop and maintain the reusability? Is it worth the trade-offs in lower tons to orbit due to more weight? Is it worth it adjusting the payload into smaller units, including developing things like refueling in LEO?
Idk, I'm not on the inside doing those calculations...
SpaceX tends to expend cores they've gotten significant use out of, rather than new ones - so the core would have been "paid off" by then.
And nasal didn't build the new rocket! They have paid Boeing 93 BILLION to design and manufacture it.
[Absolutely](https://www.reuters.com/business/finance/spacex-generated-ab...), [no way](https://payloadspace.com/estimating-spacexs-2024-revenue/), indeed.
Everything is estimated.
If you want to trust estimates and "best-guesses", neat.
SpaceX could fail tomorrow and nothing would change with national security.
national security is pretty felixaeble
Being too big to fail is not really a desirable outcome, it’s just better than failure.
Boeing is too important to fail as well but it’s been terrible as a shareholder
> > Boeing is too important to fail as well but it’s been terrible as a shareholder
Your opinion on Boeing being terrible as a shareholder vis-a-vis Tesla would be completely reversed if dividends and capital gains of the 2 companies were to be offered in the form of miles to be flown on Boeing planes and miles on Teslas Uber/Taxi/Autonomous taxis instead of dollars
The absolute overperformance on the stock market that Tesla has enjoyed vis-a-vis Boeing is not rooted in a concrete and tangible quality of life improvement for citizens. Not American citizens, nor global citizens for that matter.
It is my opinion that for all public companies in which it is possible to do so government should mandate payment in kind to all shareholders and board members to prevent the excessive promotional , cult and all around BS aspect of marketing to take over and allow people to profit just by riding off those, and Musk is the GOAT at that.
Im not comparing it to Tesla, im comparing it to any normal successful company (apple, google, nvidia, Exxon, whatever).
Boeing is an anemic company that doesn’t innovate and it should have been allowed to bankrupt and break off into businesses that worked and actually competed for customers.
Why are we still supporting this person? His cars are being outclassed internationally and he's directly meddling in this countries politics. He spectacularly failed (or wasn't it blatantly misled) the CA government with regard to the tunneling, and damaged the public sector while shutting down oversight and regulatory bodies against his companies.
Where is the benefit? These awesome tech demos? It just screams charlatan to me on an epic scale. I see no reason a government shouldn't step in to assume control if its "too big to fail".
Merging SpaceX with a public company like Tesla would create a lot of issues for the classified projects SpaceX does.
What sort of issues are you thinking?
Plenty of defense contractors with classified projects are already publicly listed, so this is not uncharted territory.
Lockhead Martin for example: https://investors.lockheedmartin.com/news-releases/news-rele...
Gives this level of detail:
> Aeronautics classified program losses $(950)
> MFC classified program losses -
It seems very safe from a national security perspective.
No? Almost every big defense contractor is publicly traded.
I imagine those are surmountable challenges. Boeing somehow manages.
But more likely that merger would consist of SpaceX acquiring Tesla and taking it private
There is no way Elon could raise the 1.4 trillion to take Tesla private
Raytheon is public.
As a SpaceX fan, I am saddened by this news.
The only reason for xAI to join SpaceX is to offload Elon's Twitter debt in the upcoming IPO.
This is just what I was thinking.
Twitter (X) owed $1.3B in debt every year in interest since Musk's takeover. This was before re-financing in a higher interest rate environment. The company was losing $200MM+ per year on ~$5B in revenue before the takeover, and there are reports that revenues have decreased by round 50%.
Best case scenario if we accept those numbers is that X makes $3B per year and about half of that goes immediately out the door in debt payments before paying a cent for the entire business to function.
However, if SpaceX acquires X, that ~$1.5B in interest is a fraction of the $8B In profits SpaceX is allegedly generating annually. Further, they can restructure the debt if it's SpaceX's debt, and not owned by X. Investors will be more likely to accept SpaceX shares as collateral than X.
> The company was losing $200MM+ per year on ~$5B in revenue before the takeover, and there are reports that revenues have decreased by round 50%.
X made a profit last year because they cut costs lower than the drop in ad revenue (which is also slowly recovering). The big question is if they will still be profitable in 2026 year without the US election driving big traffic numbers and ads.
As far as I understand they did not make a profit in 2025. They posted positive adjusted EBITDA, which is not the same.
You're right, wrote that from memory. It was EBITDA that surpassed anything Twitter previously had before purchasing it.
> Despite a revenue drop from $5 billion in 2021 to roughly $2.7 billion in 2024, the EBITDA margin surged from 13.6% to 46.3% due to drastic cost-cutting measures and restructuring
https://x.com/ekmokaya/status/1887398225881026643
How do you "know" this? They're private and don't need to report anything.
You also have to be careful about who said it and what they meant by "profit," because there is gross profit, EBIT, EBITDA, and others.
Banks released pricing to sell their debt. When the debt gets to valued near market value, it means it is essentially guaranteed to get paid back. The company was making much less money but was more profitable, see the other posters comment on EBITDA.
They report these numbers to their investors who leak them to the press.
So we come to CSAM, online nazi bar and rockets with questionable standards all mixed into one. What could possibly go wrong
As an old twitter fan, same. I was hoping elon would lose twitter to the banks in a bankruptcy.
call me crazy but if this isn't related to the AI stuff in military stuff nothing is. Anyone who believes spacex not that is naive
Either this is a straight up con, or Musk found a glitch in physics. It's extremely difficult to keep things cold in space.
He buys twitter at an inflated valuation. Runs it to the ground to a much lower valuation of $9B. [1] Then, his company Xai buys Twitter at a $33B, inflating the valuation up. Then SpaceX merges with Xai for no particular reason, but is expected to IPO at a $1T+ in the upcoming years. [3]
I’m not that smart, but if I were, I would be thinking this is an extended way to move the losses from the Twitter purchase on to the public markets.
[1] https://www.axios.com/2023/12/31/elon-musks-x-fidelity-valua...
[2] https://www.reuters.com/markets/deals/musks-xai-buys-social-...
[3] https://www.cnbc.com/amp/2026/02/02/elon-musk-spacex-xai-ipo...
It also makes it impossible for Twitter/X to die, as it deserves. It is by far the most toxic mainstream social network. It has an overwhelming amount of far right supremacist content. So bad that it literally resulted in Vivek Ramaswamy, a gubernatorial candidate in Ohio, to quit Twitter/X - nearly 100% of replies to his posts were from far right racists.
Obviously advertisers have not been fans. And it is a dying business. But rather than it dying, Elon has found a clever (and probably illegal) way to make it so that SpaceX, which has national security importance, is going to prop up Twitter/X. Now our taxpayer dollars are paying for this outrageous social network to exist.
I find HN and the tech circles to be one of the main community pillars holding up X. None of my social friends use it anymore, but links absolutely abound here, and it seems like the standard line is to pretend Elon, Grok, all the one button revenge and child porn etc don’t exist. I truly can’t fathom the amount of not thinking about it it would take to keep using the platform.
I have a blocker set up in my browser to prevent accidental clicks and sending any traffic to them when I'm not careful to check a given HN link to a posting. I've never had an account there (nor any of the popular social media networks) but I don't want to send even my few clicks their way.
Just use lists, "Your Followers" tab and never touch the "For You" tab and its basically the same as Twitter was 5 years ago.
No it isn't, the sensible people you followed 5 years ago left and stopped posting. The "Your followers" feed is now just the terminally addicted and the angry demagogues.
I am with you 100%.
It was easy to support SpaceX, despite the racist/sexist/authoritarian views of its owner, because he kept that nonsense out of the conversation.
X is not the same. Elon is actively spewing his ultraconservative views on that site.
Now that these are the same company, there's no separation. SpaceX is part of Musk's political mission now. No matter how cool the tech, I cannot morally support this company, and I hope, for the sake of society, it fails.
This announcement, right after the reveal that Elon Musk reached out to Jeffrey Epstein and tried to book a trip to Little St. James so that he could party with "girls", really doesn't bode well.
It's a shame you can't vote these people out, because I loved places like Twitter, and businesses like SpaceX and Tesla, but Elon Musk is a fascist who uses his power and influence to attack some of the most important pillars of our society.
You kinda can, just don’t make a Twitter account, don’t buy teslas, don’t use grok. Tell your friends
> X is not the same. Elon is actively spewing his ultraconservative views on that site.
I wonder if Musk would be willing to let a journalist do a deep dive on all internal communications in the same way he did when he took over twitter.
That was not a journalist.
Elon is moderate at best. If a democrat supported cutting the budget, having an actual border to the country, and keeping men out of women's bathrooms you'd get Elon.
Check out his X feed. He regularly posts unhinged things about white culture, western values, etc that are supremacist and often, lifted from other supremacists. In the last year he became far more radicalized towards the far right. If it was just the things you said I might agree.
> It also makes it impossible for Twitter/X to die, as it deserves. It is by far the most toxic mainstream social network. It has an overwhelming amount of far right supremacist content.
Twitter also has more (not total, but more) free speech than any other social networking site. For example, you are allowed to discuss empirical research on race, crime and IQ. That would get you rate limited or banned quickly on other websites, including HN.
You literally get shadowbanned for posting the three letters “cis”.
This isn't really true, though? The ISS does it with radiators that are ~1/2 the area of its solar panels, and both should scale linearly with power?
ISS radiators run on water and ammonia. Think about how much a kg costs to lift to space and you'll see the economics of space data centers fall apart real fast. Plus, if the radiator springs a leak the satellite is scrap.
The point of the Starship program is to drop the cost of a kg going to space significantly - this isn't meant to be launched with rockets that aren't fully reusable.
Even if power cost nothing the limiting factor on data center value creation is distance to where the data is requested. Putting it in space is dumb.
Not all AI workloads are latency sensitive.
The ISS creates radically less heat than a datacenter
I don't pretend to understand the thermodynamics of all of this to do an actual calculation, but note that the ISS spends half its time in the shadow of the earth, which these satellites would not do.
Wouldn't they?
You would put these in polar orbits so they are always facing the Sun. Basically the longitude would follow the Sun (or the terminator line, whichever you prefer), and the latitude would oscillate from 90°N to 90°S and back every 24 hours.
From the linked article:
> By directly harnessing near-constant solar power
Implies they would not spend half of their time in the dark.
No. Otherwise how would you power them? We could use nuclear power methods, like we did in the Voyagers for instance. But the press release doesn’t mention that and, for a constellation of satellites around the earth, it would be a terrible idea.
Radiator size scales linearly with power but, crucially, coolant power, pumps, etc do not.
Imagine the capillary/friction losses, the force required, and the energy use(!) required to pump ammonia through a football-field sized radiator panel.
Also, space solar is around 4-8x more efficient (24h/day full sun instead of ~4-8 on Earth), and 40% gain due to no atmospheric loss.
Moving electricity long distance is a lot easier than moving coolant long distances, which puts a soft limit on the reasonable size of the solar array of these satellites. But as long as you stay below that and pick a reasonable orbit it's indeed not too bad, you just have to properly plan for it
The ISS isn't consuming and generating megawatts+ of power.
Yes but if the solar panel area scales linearly with radiator area, the problem doesn't get worse?
It does if you don't turn off the heat source every 30 minutes or so. Since the "datacenters" are targeted at sun synchronous orbits they have 24/7 heat issues. And they convert pretty much all collected energy into heat as well (and some data, but that's negligible). Those GPUs are not magically not generating heat.
Wouldn't the panels themselves need cooling too? The ones on earth generate heat while being in the sun.
There are commercial systems that can use open loop cooling (i.e. spray water) to improve efficiency of the panel by keeping the panel at a optimal temp of ~25C and the more expensive closed loop systems with active cooling recovers additional energy from the heat by circulating water like a solar heater in the panel back.
I would hope SpaceX is using more efficient solar cells than the ISS
Probably not. The ISS got a solar array upgrade after its initial launch:
https://www.spectrolab.com/company.html
Twenty-five years after the ISS began operations in low Earth orbit, a new generation of advanced solar cells from Spectrolab, twice as efficient as their predecessors, are supplementing the existing arrays to allow the ISS to continue to operate to 2030 and beyond. Eight new arrays, known as iROSAs (ISS Roll-Out Solar Arrays) are being installed on the ISS in orbit.
The new arrays use multi-junction compound semiconductor solar cells from Spectrolab. These cells cost something like 500 times as much per watt as modern silicon solar cells, and they only produce about 50% more power per unit area. On top of that, the materials that Spectrolab cells are made of are inherently rare. Anyone talking about scaling solar to terawatts has to rely on silicon or maybe perovskite materials (but those are still experimental).
The physical constraints aren’t insane for black body radiators. The engineering to run radiators at 90C in space OTOH…
Setting aside the possibility it's window dressing for a financial bailout, there would be two ways compute in space makes sense:
1) new technology improves vacuum heat radiation efficiency
2) new technology reduces waste heat generation from compute
All the takes I've seen have been focused on #1, but I'm starting to wonder about #2... Specifically spintronics and photonic chips.
If you solve 2, heat dissipation goes away on earth too, so what’s the advantage of space
I'm not the best person to make that case as I can only speculate (land cost, permitting, latency, etc). /Shrug
In all the conversations I've seen play out on hacker news about compute in space, what comes up every time is "it's unviable because cooling is so inefficient".
Which got me thinking, what if cooling needs dropped by orders of magnitude? Then I learned about photonic chips and spintronics.
If you're considering only viability, the obvious concern would be cooling, yes; because increasingly large radiative cooling systems dominate launch costs because of all the liquid you need to boost into orbit. And one 100MW installation would be 500 times the largest solar power/radiative cooling system we've ever launched, which is the ISS. So get that down 2 orders of magnitude and you're within the realm of something we _know_ is possible to do instead of something we can _speculate_ is possible.
After that frankly society-destabilizing miracle of inventing competitive photonic processing, your goal of operating data centers in space becomes a tractable economic problem:
Pros:
- You get a continuous 1.37 kW/m^2 instead of an intermittent 1.0 kW/m^2
- Any reasonable spatial volume is essentially zero-cost
Cons:
- Small latency disadvantage
- You have to launch all of your hardware into polar orbit
- On-site servicing becomes another economic problem
So it's totally reasonable to expect the conversation to revolve around cooling, because we know SpaceX can probably direct around $1T into converting methane into delta-V to make the economics work, but the cooling issue is the difference between maybe getting one DC up for that kind of money, or 100 DCs.
Do you mind expanding on "society-destabalizing"?
Well, the primary limit on computation today is heat dissipation (the "power wall"). You either need to limit power so your phone or laptop doesn't destroy itself, or pay more to evacuate heat produced by the chips in your data center, which has its own efficiency curve.
If we suddenly lose 2 orders of magnitude of heat produced by our chips, that means we can fit 2 orders of magnitude more compute in the same volume. That is going to be destabilizing in some way, at the very least because you will get the same amount of compute in 1% the data center square footage of today; alternatively, you will get 100-900x the compute in today's data center footprint. That's like going from dial-up to fiber.
This is true for all of Elon's space ambition, fwiw.
> space is called “space” for a reason.
you think we don't have enough space on earth for a few buildings? this seems like a purely western cope. China seems perfectly able to build out large infrastructure projects with a land area smaller than that of the continentenal USA
> China seems perfectly able to build out large infrastructure projects with a land area smaller than that of the continentenal USA
China has a land area greater than the USA. (Continental or otherwise.)
Not true. China 9.6 million square kilometers, USA 9.8 million square kilometers, contiguous 8.1 million.
You're presumably looking at a source that's including water area. When talking about land area, China > USA > Canada. (As opposed to when including water area, Canada > USA > China)
sure, we can neglect the water, but the USA has much more usable flat land than China, and that is a pretty inarguable point.
1. It's cheaper to make a vacuum on earth around a computer than it is to send a computer into space.
2. That would also presumably work on earth, unless it somehow relied on low-gravity, and would also be cheaper to benefit from on earth.
That’s not what 1 is about.
The problem for 1 is how do you dissipate heat without being in contact with a lower temperature mass.
Creating a vacuum on earth would solve nothing as the heath would still have to escape the vacuum.
> new technology improves vacuum heat radiation efficiency
Isn't this fixed by blackbody radiation equations?
That equation have surface area ? What if new material found to be extremely large surface area to weight ratio to dissipate lots of heat ?
Maybe Karpathy has been hired to design a Full Self Cooling system.
It's a con, his AI business is failing, so he's rolling it up into the profitable business. Did a similar thing with Twitter.
This is so obvious, but it's so stupid and at this scale that people find it hard to believe.
If they AI business is failing why did they just do a successful large raise?
If the AI business is successful why does it keep changing hands? Where are the profits?
Existing satellites manage to keep their equipment that already can consume several kW cool just fine.
You might need space for radiators, but there is plenty space in space.
5,000 Starship launches to match the solar/heat budget of the 10GW "Stargate" OpenAI datacenter. The Falcon 9 family has achieved over 600 launches.
The ISS power/heat budget is like 240,000 BTU/hr. That’s equivalent to half of an Nvidia GB200 NVL72 rack. So two international space stations per rack. Or about 160,000 international space stations to cool the 10GW “Stargate” datacenter that OpenAI’s building in Abilene. There are 10,000 starlink satellites.
Starship could probably carry 250-300 of the new V2 Mini satellites which are supposed to have a power/heat budget of like 8kW. That's how I got 5,000 Starship launches to match OpenAI’s datacenter.
Weight seems less of an issue than size. 83,000 NVL72’s would weigh 270 million lbs or 20% of the lift capacity of 5000 starship launches. Leaving 80% for the rest of the satellite mass, which seems perhaps reasonable.
Elon's napkin math is definitely off though, by over an order of magnitude. "a million tons per year of satellites generating 100 kW of compute power per ton" The NVL72's use 74kW per ton. But that's just the compute, without including the rest of the fucking satellite (solar panels and radiators). So that estimate is complete garbage.
One note: If you could afford to send up one of your own personal satellites, it would be extremely difficult for the FBI to raid.
Several kW is nothing for a bank of GPUs.
Radiators in space are extremely inefficient because there's no conduction.
Also you have huge heat inputs from the sun. So you need substantial cooling before you get around to actually cooling the GPUs.
you put the radiators and the rest of the satellite within the shade of the solar panels, you can still make the area arbitrarily large
EDIT: people continue downvoting and replying with irrelevant retorts, so I'll add in some calculations
Let's assume
1. cheap 18% efficient solar panels (though much better can be achieved with multijunction and quantum-cutting phosphors)
2. simplistic 1360 W/m^2 sunlight orthogonal to the sun
3. an abstract input Area Ain of solar panels (pretend its a square area: Ain = L ^ 2)
4. The amount of heat generated on the solar panels (100%-18%) * Ain * 1360 W / m ^ 2, the electrical energy being 18% * Ain * 1360 W / m ^ 2. The electrical energy will ultimately be converted to computational results and heat by the satellite compute. So the radiative cooling (only option in space) must dissipate 100% of the incoming solar energy: the 1360 W / m^2 * Ain.
5. Lets make a pyramid with the square solar panel as a base, with the apex pointing away from the sun, we make sure the surface has high emissivity (roughly 1) in thermal infrared. Observe that such a pyramid has all sides in the shade of the sun. But it is low earth orbit so lets assume warm earth is occupying one hemisphere and we have to put thermal IR reflectors on the 2 pyramid sides facing earth, so the other 2 pyramid sides face actual cold space.
6. The area for a square based symmetric pyramid: we have
6.a. The area of the base Ain = L * L.
6.b. The area of the 4 sides 2 * L * sqrt( L ^ 2 / 4 + h ^ 2 )
6.c. The area of just 2 sides having output Area Aout = L * sqrt( L ^ 2 / 4 + h ^ 2 )
7. The 2 radiative sides not seeing the sun and not seeing the earth together have the area in 6.c and must dissipate L ^ 2 * 1360 W / m ^ 2 .
8. Hello Stefan-Boltzmann Law: for emissivity 1 we have the radiant exitance M = sigma * T ^ 4 (units W / m ^ 2 )
9. The total power exited through the 2 thermal radiating sides of the pyramid is then Aout * M
10. Select a desired temperature and solve for h / L (to stay dimensionless and get the ratio of the pyramid height to its base side length), lets run the satellite at 300 K = ~26 deg C just as an example.
11. If you solve this for h / L we get: h / L = sqrt( ( 1360 W / m ^ 2 / (sigma * T ^ 4 ) ) ^ 2 - 1/4 )
12. Numerically for 300K target temperature we get: h/L = sqrt((1360 / (5.67 * 10^-8 * 300 ^ 4)) ^ 2 - 1/4) = 2.91870351609271066729
13. So the pyramid height of "horribly poor cooling capability in space" would be a shocking 3 times the side length of the square solar panel array.
As a child I was obsessed with computer technology, and this will resonate with many of you: computer science is the poor man's science, as soon as a computer becomes available in the household, some children autodidactically educate themselves in programming etc. This is HN, a lot of programmers who followed the poor man's science path out of necessity. I had the opportunity to choose something else, I chose physics. No amount of programming and acquiring titles of software "engineer" will be a good substitute for physicists and engineers that actually had courses on the physical sciences, and the mathematics to follow the important historical deductions... It's very hard to explain this to the people who followed the path I had almost taken. And they downvote me because they didn't have the opportunity, courage or stamina to take the path I took, and so they blindly copy paste each others doomscrolled arguments.
Look I'm not an elon fanboy... but when I read people arguing that cooling considerations excludes this future, while I know you can set the temperature arbitrarily low but not below background temperature of the universe 4 K, then I simply explain that obviously the area can be made arbitrarily large, so the temperature can be chosen by the system designer. But hey the HN crowd prefers the layers of libraries and abstractions and made themselves an emulation of an emulation of an emulation of a pre-agreed reality as documented in datasheets and manuals, and is ultimately so removed from reality based communities like physics and physics engineering, that the "democracy" programmers opinions dominate...
So go ahead and give me some more downvotes ;)
If you like mnemonics for important constants: here's one for the Stefan Boltzman constant: 5.67 * 10^-8 W / m^2 / K ^ 4
thats 4 consecutive digits 5,6,7,8 ; comma or point after the first significant digit and the exponent 8 has a minus sign.
It's really not that simple. See this for a good explanation of why: https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
It all basically boils down to: in order to dissipate heat, you need something to dissipate heat into, e.g. air, liquid, etc. Even if you liquid cool the GPUs, where is the heat going to go?
On Earth, you can vent the heat into the atmosphere no problem, but in space, there's no atmosphere to vent to, so dissipating heat becomes a very, very difficult problem to solve. You can use radiators to an extent, but again, because no atmosphere, they're orders of magnitude less effective in space. So any kind of cooling array would have to be huge, and you'd also have to find some way to shade them, because you still have to deal with heat and other kinds of radiation coming from the Sun.
It's easier to just keep them on Earth.
What you're describing is one of two mechanisms of shedding heat which is convection, heating up the environment. What the long comment above is describing is a _completely_ different mechanism, radiation, which is __more__ efficient in a vacuum. They are different things that you are mixing up.
for a square solar array of side length L, a pyramid height of 3*L would bring the temperature to below 300K, check my calculation above.
people heavily underestimate radiative cooling, probably because precisely our atmosphere hinders its effective utilization!
lesson: its not because radiative cooling is hard to exploit on earth at sea level, that its similarily ineffective in space!
that page has not a single calculation of radiative heat dissipation, seems like he pessimistically designed the satellite avoiding use of radiative cooling which forces him to employ a low operational duty cycle. Kind of a shame to be honest, given the high costs of launching satellites, his sat could have been on for a larger fraction of time...
It seems straightforward to you because you're ignoring everything that makes this not work.
Here's a big one: you can't put radiators in shadow because the coolant would freeze. ISS has system dedicated to making sure the radiators get just enough sunlight at any given time.
since you make the same argument in 2 places, I will refer you to my response in the other place you made the same argument:
https://news.ycombinator.com/item?id=46867514
That helps with the heat from the sun problem, but not the radiation of heat from the GPUs. Those radiators would need to be unshaded by the solar panels, and would need to be enormous. Cooling stuff in atmosphere is far easier than in vacuum.
Not so. Look at the construction of JWST. One side is "hot", the other side is very, very cold.
I am highly skeptical about data centers in space, but radiators don't need to be unshaded. In fact, they benefit from the shade. This is also being done on the ISS.
That's fair. I meant they would need a clear path to open space not blocked by solar panels, but yes, a hot and cold side makes sense.
The whole concept is still insane though, fwiw.
"I meant they would need a clear path to open space not blocked by solar panels, but yes, a hot and cold side makes sense."
This is precisely why my didactic example above uses a convex shape, a pyramid. This guarantees each surface absorbs or radiates energy without having to take into account self-obscuring by satellite shape.
Look at how many layers of insulation are needed for the JWST to have a hot and cold side! Again, this is not particularly simple stuff.
The JWST operates at 2kw max. That's not enough for a single H200.
AI datacenters in space are a non-starter. Anyone arguing otherwise doesn't understand basic thermodynamics.
The goal of JWST is not to consume as much power as possible, and perform useful computations with it. A system not optimized for metric B but for metric A scores bad for metric B... great observation.
> Look at the construction of JWST.
A very high end desktop pulls more electricity than the whole JWST... Which is about the same as a hair dryer.
Now you need about 50x more for a rack and hundreds/thousands racks for a meaningful cluster. Shaded or not it's a shit load of radiators
https://azure.microsoft.com/en-us/blog/microsoft-azure-deliv...
addressed at https://news.ycombinator.com/item?id=46867402
this makes no sense, the radiation of heat from the GPU's came from electrical energy, the electrical energy came from the efficient fraction of solar panel energy, the inefficient fraction being heating of the solar panel, the total amount of heat that needs to be dissipated is simply the total amount of energy incident on the solar panels.
True, the solar panels would need to be enormous too.
Let's say we wanted to train LLaMa 3.1 405B:
[0] https://developer.nvidia.com/deep-learning-performance-train...
Click the "Large Language Model" tab next to the default "MLPerf Training" tab.
That takes 16.8 days on 128 B200 GPU's:
> Llama3 405B 16.8 days on 128x B200
A DGX B200 contains 8xB200 GPU's. So it takes 16.8 days on 16 DGX B200's.
A single DGX (8x)B200 node draws about 14.3 kW under full load.
> System Power Usage ~14.3 kW max
source [1] https://www.nvidia.com/en-gb/data-center/dgx-b200
16 x 14.3 kW = ~230 kW
at ~20% solar panel efficiency, we need 1.15 MW of optical power incident on the solar panels.
The required solar panel area becomes 1.15 * 10^6 W / 1.360 * 10^3 W / m ^ 2 = 846 m ^ 2.
thats about 30 m x 30 m.
From the center of the square solar panel array to the tip of the pyramid it would be 3x30m = 90 m.
An unprecedented feat? yes. But no physics is being violated here. The parts could be launched serially and then assembled in space. Thats a device that can pretrain from scratch LLaMa 3.1 in 16.8 days. It would have way to much memory for LLaMa 3.1: 16 x 8 x 192 GB = ~ 25 TB of GPU RAM. So this thing could pretrain much larger models, but would also train them slower than a LLaMa 3.1.
Once up there it enjoys free energy for as long as it survives, no competing on the electrical grid with normal industry, or domestic energy users, no slow cooking of the rivers and air around you, ...
> arbitrarily large
Space is not empty. Satellites have to be boosted all the time because of drag. Massive panels would only worsen that. Once you boosters are empty the satellite is toast.
the point wasn't that a 1 m^2 solar panel could theoretically be kept reasonably cool at the cost of a miles long radiator... nono, the point was that you could attain any desirable temperature this way, arbitrarily close to 4K.
for a reasonable temperature (check my comment for updated calculations) the height of a square based pyramidal satellite would be about 3 times the side length of its base, quite reasonable indeed. Thats with the square base of the pyramid as solar panel facing the sun, and the top of the pyramid facing away, so all sides are in the shade of the base. I even halved my theoretical cooling power to keep calculations simple: to avoid a long confusing calculation of the heat emitted by earth, I handicapped my design so 2 of the pyramidal side surfaces are reflective (facing earth) and the remaining 2 side triangles of the pyramid are the only used thermal radiative cooling surfaces. Less pessimistic approaches are possible, but would make the calculation less didactic for the HN crowd.
It seems straightforward to you because you're ignoring everything that makes this not work.
Here's a big one: you can't put radiators in shadow because the coolant would freeze. ISS has system dedicated to making sure the radiators get just enough sunlight at any given time.
The ISS goes into Earth's shadow for ~45 minutes and then in the sun for 45 minutes, in 24/7 repeat;
this system would not be given such an orbit. Its trivial to decrease the cooling capacity of the radiators: just have an emissivity ~0 shade (say an aluminum foil) curtain obscure part of the radiator so that it locally sees itself instead of cold empty space. This would only happen during 2 short periods in the year.
The design issues of the ISS are totally different from this system.
"Satellites have to be boosted all the time because of drag."
On Low Earth Orbits (LEOs), sure, but the traces of atmosphere that cause the drag disappear quite fast with increasing altitude. At 1000 km, you will stay up for decades.
arbitrarily large means like measured in square km. Starcloud is talking about 4km x 4km area of solar panels and radiative cooling. (https://blogs.nvidia.com/blog/starcloud/)
Building this is definitely not trivial and not easy to make arbitrarily large.
When a physicist says arbitrarily large it could even be in a dimensionless sense. It doesn't matter how small or large the solar panel is:
for a 4 m x 4 m solar panel, the height of the pyramid would have to be 12 m to attain ~ 300 K on the radiator panels. Thats also the cold side for your compute.
for a 4 km x 4 km solar panel the height of the pyramid would be 12 km.
A size like that is going to be completely, absolutely obliterated by micrometeor collisions.
These people are all smoking crack.
I’ve got a perpetual motion machine to sell you.
this isn't even an argument?
> you put the radiators and the rest of the satellite within the shade of the solar panels, you can still make the area arbitrarily large
The larger you make the area, the more solar energy you are collecting. More shade = more heat to radiate. You are not actually making the problem easier.
no the radiator planes are in the shade, so you can increase the height of a pyramidal shaped satellite for a constant solar panel base, and thus enjoy arbitrarily low rest temperatures, check my calculation which I added.
for a target temperature of 300K that would mean the pyramid height would be a bit less than 3 times higher than the square base side length h=3L.
I even handicapped my example by only counting heat radiation from 2 of the 4 panels, assuming the 2 others are simply reflective (to make the calculation of a nearby warm Earth irrelevant).
I'm not big on this subject, but I understand that heat transfer is difficult in space, because there's little to transfer to. If the solution is just making large radiators, then that means you're sending some big payloads full of radiators. Not to mention all the solar panels needed. I wanna live in sci-fi land too, but I don't see how it makes any sense compared to a terrestrial data center.
the radiators would be lighter compared to the solar panels, and slightly smaller surface area so you can line them back to back
If someone has a design out there where this works and you can launch it economically on a rocket today, I wanna see that. And then I wanna compare it to the cost of setting up some data centers on earth (which BTW, you can service in real time, it sounds like these will be one-and-done launches).
> keep their equipment that already can consume several kW cool just fine
That's equivalent to a couple datacenter GPUs.
> You might need space for radiators, but there is plenty space in space.
Finding space in space is the least difficult problem. Getting it up there is not easy.
You can line the solar panels and radiators facing away from each other, and the radiators would take up less surface area. I think maybe the tricky part would be the weight of water + pipes to move heat from the compute to the radiators.
Water is not needed to move heat. Heat pipes do it just fine. There's one in your laptop and one in your phone too. It does scale up.
Interesting, That could surely simplify things.
There's plenty of space in space, but there isn't plenty of space in rocket fairings, nor is there plenty of lift capacity for an unlimited amount of radiators.
> It's extremely difficult to keep things cold in space.
This is one of those things that's not obvious till you think about it.
It's such bullshit that we've decided this moron and others in his cohort can unilaterally reallocate such vast portions of humanity's labor at their whims.
This is an extremely stupid idea, but because of our shared delusion of capitalism and the idea that wealth accumulation at the top should be effectively limitless, this guy gets to screw around and divert actual human labor towards insane and useless projects like this rather than solving real world problems.
I wanted to come and express this thought, but you did that already very well, thanks for that.
I am saddened too by the fact that the system is designed so that people like him can waste a large amount of economic and human capital.
Just put a fan in a window.
what makes you believe this?
radiators can be made as long as desirable within the shade of the solar panels, hence the designer can pracitically set arbitrarily low temperatures above the background temperature of the universe.
Radiators can shadow each other, so that puts some kind of limit on the size of the individual satellite (which limits the size of training run it can be used for, but I guess the goal for these is mostly inference anyway). More seriously, heat conduction is an issue: If the radiator is too long, heat won't get from its base to its tip fast enough. Using fluid is possible, but adds another system that can fail. If nothing else, increasing the size of the radiator means more mass that needs to be launched into space.
please check my didactic example here: https://news.ycombinator.com/item?id=46862869
"Radiators can shadow each other," this is precisely why I chose a convex shape, that was not an accident, I chose a pyramid just because its obvious that the 4 triangular sides can be kept in the shade with respect to the sun, and their area can be made arbitrarily large by increasing the height of the pyramid for a constant base. A convex shape guarantees that no part of the surface can appear in the hemispherical view of any other part of the surface.
The only size limit is technological / economical.
In practice h = 3xL where L was the square base side length, suffices to keep the temperature below 300K.
If heat conduction can't be managed with thermosiphons / heat pipes / cooling loops on the satellite, why would it be possible on earth? Think of a small scale satellite with pyramidal sats roughly h = 3L, but L could be much smaller, do you actually see any issue with heat conduction? scaling up just means placing more of the small pyramidal sats.
Shading does work; JWST does this. However I don't see how you can make it work for satellite data centers. You would constantly be engaging attitude control as you realigned the panels to keep the radiators in shade. You'd run out of thruster fuel so fast you'd get like a month out of each satellite
attitude control doesn't need to consume propellant, there's reaction wheels.
but you'd rarely ever need it though: it just needs to rotate at a low angular velocity of 1 rotation per year to keep facing the sun.
https://en.wikipedia.org/wiki/Spacecraft_attitude_determinat...
Radiators can only be made as long as desirable because there's gravity for the fluid inside to go back down once it condenses. Even seen those copper heat pipes in your PC radiator?
Fluid in heat pipes moves through capillary action.
these same comments pop up every time someone brings up satellite data-centers where people just assume the only way of dissipating heat is through convection with the environment.
No, we just "assume" (i.e. know) that radiation in a vacuum is a really bad way of dissipating heat, to the point that we use vacuum as a very effective insulator on earth.
Yes, you can overcome this with enough radiator area. Which costs money, and adds weight and space, which costs more money.
Nobody is saying the idea of data centers in space is impossible. It's obviously very possible. But it doesn't make even the slightest bit of economic sense. Everything gets way, way harder and there's no upside.
> No, we just "assume" (i.e. know) that radiation in a vacuum is a really bad way of dissipating heat, to the point that we use vacuum as a very effective insulator on earth.
In space or vacuum radiation is the best way to dissipate heat, since it's the only way.
I believe the reason the common person assumes thermal radiation is a very poor way of shedding heat is because of 2 factoids commonly known:
1. People think they know how a vacuum flask / dewar works.
2. People understand that in earthly conditions (inside a building, or under our atmosphere) thermal radiation is insignificant compared to conduction and convection.
But they don't take into account that:
1) Vacuum flasks / dewars use a vacuum for thermal insulation. Yes and they mirror the glass (emissivity nearer to ~0) precisely because thermal radiation would occur otherwise. They try their best to eliminate thermal radiation, a system optimized to eliminate thermal radiation is not a great example of how to effectively use thermal radiation to conduct heat. The thermal radiation panels would be optimized for emissivity 1, the opposite of whats inside the vacuum flask.
2) In a building or under an atmosphere a room temperature object is in fact shedding heat very quickly by thermal radiation, but so are the walls and other room temperature objects around you, they are reheating you with their thermal radiation. The net effect is small, in these earthly conditions, but in a satellite the temperature of the environment faced by the radiating surfaces is 4K, not a temperature similar to the object you are trying to keep cool.
People take the small net effect of thermal radiation in rooms etc, and the slow heat conduction through a vacuum flasks walls as representative for thermal radiation panels facing cold empty space, which is the mistake.
Well no, it’s because conduction/convection into a fluid is so much more effective.
Just look at a car. Maybe half a square meter of “radiator” is enough to dissipate hundreds of kW of heat, because it can dump it into a convenient mass of fluid. That’s way more heat than the ISS’s radiators handle, and three orders of magnitude less area.
Or do a simple experiment at home. Light a match. Hold your finger near it. Then put your finger in the flame. How much faster did the heat transfer when you made contact? Enough to go from feeling mildly warm to causing injury.
Yes, it's so much more effective, ... at sea level Earthly conditions.
What’s more effective: conduction/convection on the ground, or radiation in space?
but thats what you don't get: conduction / convection on the ground is ultimately still radiation to space: you heat up our rivers, soils, atmosphere and the heat is eventually shed... by thermal radiation.
its not exactly good advertisement for conductive or convective heat transfer if its really employing thermal radiation under the hood!
but do you want big tech to shit where you eat? or do you want them to go to the bathroom upstairs?
At some point I'm thinking the large resistance to the idea I am seeing in a forum populated with programmers is the salivation-inducing idea that all that datacenter hardware will eventually get sold for less and less, but if we launch them to space there won't be any cheap devalued datacenter hardware to put in their man-caves.
You have presented a good case from the physics textbook for calculating the radiator size.
However, what do you reckon the energy balance is for launching the 1 GW datacenter components into space and assembling it?
Additional radiator area means bigger spacecraft, implies more challenge with attitude control. Lower down you get more drag so you use propellant to keep yourself up, higher up you have more debris and the large area means you need to frequently manoeuvre to avoid collisions. Making things bigger in space is not trivial! You can't just deploy arbitrarily large panels and expect everything to be fine.
space is vast
they could go near a Lagrange point
there are so many options
heavier boats are also slower to accelerate or decelerate compared to smaller boats, does this mean we should ban container ships? having special orbits for megastructure lanes would seem a reasonable approach.
The radiators would be lighter compared to the solar panels, and slightly smaller surface area so you can line them back to back
I don't think dissipating heat would be an issue at all. The cost of launch I think is the main bottleneck, but cooling would just be a small overhead on the cost of energy. Not a fundamental problem.
If you solved this problem apply at nasa because they still haven't figured it out.
Either that or your talking out of your ass.
FYI a single modern rack consumes twice the energy of the entire ISS, in a much much much much smaller package and you'll need thousands of them. You'd need 500-1000 sqm of radiator per rack and that alone would weight several tonnes...
You'll also have to actively cool down your gigantic solar panel array
eldenring is slightly wrong: for reasonable temperatures the area of the radiating panels would have to be a bit more than 3 times the area of the solar panel, otherwise theres nothing wrong.
No need to apply at NASA, to the contrary, if you don't believe in Stefan Boltzmann law, feel free to apply for a Nobel prize with your favorite crank theory in physics.
Whats your definition for reasonable temp? my envelope math tells me at 82 celsius (right before h100s start to throttle) you'd need about 1.5x the surface area for radiators. Not exactly back to back, but even 3x surface area is reasonable.
Also this assumes a flat surface on both sides. Another commenter in this thread brought up a pyramid shape which could work.
Finally, these gpus are design for earth data centers where power is limited and heat sinks are abundant. In the case of space data centers you can imagine we get better radiators or silicon that runs hotter. Crypto miners often run asics very hot.
I just don't understand why every time this topic is brought up, everyone on HN wants to die on the hill that cooling is not possible. It is?? the primary issue if you do the math is clearly the cost of launch.
I am the person who gave the pyramid shape as a didactic example (convexity means we can ignore self obscuration, and giving up 2 of the 4 triangular side surfaces of the pyramid allows me to ignore the presence of lukewarm earth).
My example is optimized not for minimal radiator surface area, but for minimal mathematical and physical knowledge required to understand feasibility.
Your numbers are different because you chose 82 C (355 K) instead of my 26 C (300 K).
Near normal operating temperatures hardware lifetime roughly doubles for every 10 deg C/K decrease in temperature (this does not hold indefinitely of course).
You still need to move the heat from the GPU to the radiator so my example of 26 deg C at the radiator just leaves a lot of room against criticism ;)
The pertinent thing is that it’s not an advantage. It may be doable but it’s not easier than cooling a computer in a building.
The distinction is that you don't need to compete for land area, that you don't cause local environmental damage by heating say a river or a lake, that you don't compete with meatbags for energy and heat dissipation rights.
Without eventually moving compute to space we are going to have compute infringe on the space, energy, heat dissipation rights of meatbags. Why welcome that?!?
How efficient is thermal radiation through a vacuum again?
Sure, it occurs, but what does the Stefan–Boltzmann law tell us about GPU clusters in space?
> How efficient is thermal radiation through a vacuum again?
I provided the calculation for the pyramidal shape: if the base of a pyramid were a square solar panel with side length L, then for a target temperature of 300K (a typical back of envelope substitute for "room temperature") the height of the pyramid would have to be about 3 times the side length of the square base. Quite reasonable.
> Sure, it occurs, but what does the Stefan–Boltzmann law tell us about GPU clusters in space?
The Stefan-Boltzmann law tells us that whatever prevents us from putting GPU clusters in space, it's not the difficulty in shedding heat by thermal radiation that is supposedly stopping us.
Is it the required size of the wings for radiative cooling then?
Just picture a square based pyramid, like a pyramid from egypt, thats the rough shape. Lets pretend the bottom is square. For thermodynamic analysis, we can just pretend the scale is irrelevant, it could be 4 cm x 4 cm base or 4 km x 4 km base. Now stretch the pyramid so the height of the tip is 3 times the length of the sides of the square base, so 12 cm or 12 km in the random examples above.
If the base were a solar panel aimed perpendicular to sun, then the tip is facing away and all side triangles faces of the pyramid are in the shade.
I voluntarily give up heat dissipation area on 2 of the 4 triangular sides (just to make calculations easier, if we make them thermally reflective -emissivity 0-, we can't shed heat, but also don't absorb heat coming from lukewarm Earth).
The remaining 2 triangular sides will be large enough that the temperature of the triangular panels is kept below 300 K.
The panels also serve as the cold heat baths, i.e. the thermal sinks for the compute on board.
Not sure what you mean with wings, I intentionally chose a convex shape like a pyramid so that no part of the surface of the pyramid can see another part of the surface, so no self-obstruction for shedding heat etc...
If this doesn't answer your question, feel free to ask a new question so I understand what your actual question is.
The electrical power available for compute will be approximately 20% (efficiency of solar panels) times the area of the square base L ^ 2 times 1360 W / m ^ 2 .
The electrical power thus scales quadratically with the chosen side length, and thus linearly with the area of the square base.
what? the heat is coming from inside the house
which house?
This is why I come to this site. Obviously, Twitter's financials are struggling and theres more than a few people rich people who don't want to take the hit... but we can all drop that for a second to discuss the plausibility of data centers in space. Some links and comments I enjoyed:
The Goalpost shift continues, If elon were working for me, I would have fired him for having never delivered on any of his projects.
Hyperloop > Neuralink > Self-Driving Cars > Robotaxi fleets > Personal Robots > Orbital Datacenters > [Insert next]
> If elon were working for me, I would have fired him for having never delivered on any of his projects.
Never? For the sheer amount of moonshot bets he's doing, his track record would make any VC jealous. Zip2, PayPal, Tesla, SpaceX, Grok/xAI.
World's Best At Surfing A Temporary Hyperinflation Wave is not a life goal to really be proud of tbh.
Plenty of politicians are very proud of doing that.
>Zip2
I guess props to scamming Compaq into making a large investment that didn't pan out. He did personally make money so I guess win for him.
>In an effort to woo investors, Elon Musk built a large casing around a standard computer to give the impression that Zip2 was powered by a supercomputer.
>PayPal
Huh? He didn't found Paypal, his company was acquired by Paypal. You might as well give him credit for eBay while you're at it. Paypal released their first digital wallet in 1999. They acquired x.com (and Musk) in 2000. Paypal itself was then acquired by eBay in 2002.
>Tesla
Investor, not founder.
>SpaceX
Yup, props here.
>Grok/xAI
Hasn't made a penny, no signs it had any path to profitability, which is why it was shoved into Space-X to cover his personal losses.
Taking Tesla from where it was (an overpriced prototype) to what it is now did take some skill. He wasn't some passive investor who put money in and didn't do anything. The rest for sure he was gotten credit that isn't earned.
And of course, Elon did all of that work personally. 16 hours a day, nonstop hard work. Just take his word for it, it means a lot.
hey now. Lets not forget when Elon had Grok creating CSAM and sexually explicit material of nonconsenting women. Truely an... achievement? Surely it will propel humanity forward.
note the post you're replying to said "track record" not "founded"
Hey now. Huperliop was designed to scuttle California’s light rail project. Which it did. Mission accomplished.
I think you mean "California High-Speed Rail", not light rail.
Light rail, generally refers to urban rail, "trams".
Still failed and missed every projection. I think he said we would be on Mars too?
California's light rail still exists, as does California High-Speed Rail, which is not light rail.
This is the funniest conspiracy theory.
It’s not really so much a conspiracy theory as a thing that he outright said.
https://www.jalopnik.com/did-musk-propose-hyperloop-to-stop-...
He can take out a full page Wall Street Journal ad tomorrow that says “I created hyperloop to kill CA HSR” and it will have no effect on the fact that CA HSR’s failure is 100% the fault of CA’s own dysfunction.
Yeah that’s where I’m confused about this “conspiracy theory” stuff. It’s common knowledge that Musk wanted hyperloop to undermine the high speed rail project and also it later failed. Aside from a single HN comment I have never seen anyone attribute him with that much influence on the thing, so it is bizarre to see someone talking like there’s some sort of common conspiracy theory that Elon Musk controls trains or whatever. As far as I know pretty much nobody believes that.
That link denies the conspiracy theory?
> “Or did he just have an idea and blurt it out," I asked Vance. > "I'm 99.9-percent sure it's the latter," Vance tells me.
Also that to scapegoat Musk for killing the California train when California was perfectly able to kill it itself:
> Vance then brought up a valid point: "In all this time we've been talking about high-speed rail, there's still almost none that's built....
There is no conspiracy theory, that aside the link does not indicate that there is one? “Vaguely accurate” does not mean “untrue”, and Vance is clear that he is talking about his personal interpretation of what Elon Musk is documented to have said, which he does not refute.
I like the idea that “he didn’t say that” and “he did say that but a different guy feels like he probably meant something else” are so obviously equivalent that skepticism of that notion constitutes a ‘conspiracy theory’.
That aside I like that the guy whose opinion should be treated as indisputable fact said that he thinks that there hasn’t been any high speed rail built globally in the past decade, which is not even remotely true. Obviously if he meant to say in the US he would have said so, since his next sentence was praise of Musk’s world-wide achievements.
I suppose it’s possible that Vance either doesn’t know anything about high speed rail or was in such a rush to extoll the virtues of the CEO of Tesla that he just sort of blurted something out to make Musk look good?
> “Vaguely accurate” does not mean “untrue”
The full quote is “vaguely accurate but a disingenuous take”. And “Disingenuous” means “misleading/dishonest/untruthful/insincere/unfair”.
> Obviously if he meant to say in the US he would have said so
Come on, from the context it is clear that Vance means the US and specifically California. He also says “we” in the sentence “In all this time we've been talking about high-speed rail” and does not mean Chinese/Japanese/French having this discussion.
Disingenuous speaks to the motivation of the speaker, not the veracity of information on its own. Vance says that in his opinion that that particular interpretation of the factual information is disingenuous. As you pointed out, it can mean “unfair” which is not the same thing as untrue. Dude had an opportunity to say “that’s not true” and didn’t do that.
You’ve sort of just added “I feel like Vance meant something other than what he said” on top of Vance saying he felt like Musk meant something other than what he said. There isn’t a number of layers of “I feel…/he feels…” that you can pile onto a statement that will equal “he did not say the thing that he is quoted as having said”
Your contention is that by “accurate” he meant “inaccurate” and that he sees Elon Musk as being a global phenomenon and high speed rail as a… thing that’s local to the US? That is notable for its… absence?
Seems like “yeah that’s what he said but in my opinion you’re being mean to my friend” is more likely than a professional writer not knowing how to say “that’s not true”
It is patently clear what Musk meant, the guy isn’t famous for nuance. That aside I don’t find it difficult to picture the man that publicly claims that he personally elected the president thinking that he could sabotage a rail project. Now, I can’t know for sure that he believes that his Hyperloop pitch was responsible for the failure of the CA high speed rail project but if I had to make a bet about that…
What are you talking about?
hmm Tesla shipped millions of cars SpaceX launches 90% of space payloads, Starlink is working well. Thats hard to categorize as never delivered on any of his projects
The crucial thing is that Tesla's valuation has the hype projects baked in. The fact that it never delivered self driving or a robotaxi fleet and is now being saved solely by an import ban on Chinese EVs means that any success he had with Tesla is now an illusion.
Yep but again that does not qualify as never delivered on anything.
I don't follow his promises but have seen first hand how far ahead Tesla FSD is compared to competitors in the consumer space. It's not even close.
This current announcement seems silly, though.
Have you tested Supercruise?
There is another way to view this. FSD plays fast and loose because they are constantly iterating. The culture at Musk co is that if you dont' keep pushing updates you are in trouble so do we really want to trust that each of his numerous updates are truly tested? This guy is a pathological liar after all. How many lawsuits are they dealing with now?
Supercruise only runs on pre mapped routes. If my life is on the line, I'd rather take the pre mapped routes and supercruise design is better at preventing people playing games to defeat the system (ex.shoving an orange in the steering wheel) so I know that others using the system on the road are following the system guidelines.
Supercruise may not do everything FSD does but it cuts out a large portion of the "fatigue" portion of driving and as a result can be highly trusted value add.
> The fact that it never delivered self driving or a robotaxi fleet a
Once again pointing out Tesla has around 300 robotaxis running in 2 cities (Austin/SF).
The Starlink achievement alone is one of the most insane projects ever attempted and works really well.
The projects promised to be life altering for all mankind, they ended up being not even life altering for super rich Americans considering that Teslas are just EVs which without FSD are just regular cars with a different propellent that were made for political purposes and virtue signaling
The EV revolution has always been something almost dystopic : Trillions of dollars spent in order to not have the slightest amount of quality of life improvement, if anything a worse quality of life because you buy an EV that you cannot use 24/7/365 whereas you can an ICE car for much less .
As soon as something kinda elegant and hopeful as far as collective quality of life improvement is concerned (AI/ChatGPT) came around.....the whole green/EV revolution rightfully went out the window
If Musk was this genius you guys make him to be at 50 and with all the capital he burned he should have at least one company that if you disappeared the world would look drastically different, like if you disappeared Microsoft or Apple or Exxon or Aramco or Amazon or IBM....the world would come to a screeching halt.
Disappear one of Musk companies and everything would be the same as he's always involved in these sort of aspirational companies which have this great vision always 5 years into the future that never materialize into anything tangible or that improves the quality of life like the company I mentioned earlier
well Tesla did jump start the EV revolution not life altering but is pretty important. IF SpaceX gets spaceship right that will be a huge leap forward.
Sir, your comment appears to qualify as "moving the goal post". TSLA never delivered a single inexpensive electric vehicle, and just last week abandoned all high-end efforts (S/X/CT discontinued). All TSLA manufactures now are overpriced "meh" transport boxes. Yes, TSLA was early, and now they are far, far behind the competition.
Can we evaluate based on the stated goals, or why does the criteria keep shifting?
Tesla's goal was to accelerate the world's transition to sustainable energy by building a comprehensive ecosystem of electric vehicles (EVs), solar generation, and battery storage.
Looks to me they delivered on 2 of the 3
And now Tesla is hindering that transition.
I'm amazed at this kind of thinking. I get it, obviously, and it's not uncommon, but still.
Elon Musk has already revolutionized three industries:
1. EVs: Before Tesla, no one thought electric cars could be a mass-market product. And even today, the Model 3 and Model Y are at the top of almost all sales lists.
2. Orbital Launch: No one expected Space X to succeed. What does a software guy know about real engineering? But today, re-usable rockets are the way of the future, and Space X is at least 5 to 10 years ahead of any other company.
3. Satellite Communications: Every single major military power is trying to deploy their own version of Starlink. Before Starlink, 50 satellites was considered a big constellation. Starlink has 8,000 satellites and they are literally launching hundreds every month.
I know it's impossible to prove a counter-factual, but I'm convinced that none of these three would have happened without Elon. No other Western car company has (even now) produced a profitable EV. No other space company has prices as low as Space X. No one even has the capability to build a Starlink competitor (not yet at least). Without Elon pushing these projects, they simply would not have happened or would have happened decades later (after China or someone else beat us to it).
Even his not-yet-successful projects are far beyond most other companies:
Boring Company has actually built tunnels and passengers are actually riding it. No one else is even trying.
Neuralink has actually helped patients.
Tesla FSD actually does work (I use it all the time), and even if Waymo is ahead, Tesla is easily in second place.
I 100% get the hatred for Elon Musk. His political positions are absolutely worth criticizing and I cringe most of the time he tweets. But to deny his business and engineering ability is just motivated reasoning.
Such illusions are ultimately self-defeating. The more opposed one is to Elon Musk (in business or politics) the more important it is to see his capabilities clearly.
> Boring Company has actually built tunnels and passengers are actually riding it. No one else is even trying.
Boring Company bought an existing tunnel boring machine (TBM), and used it to dig a car tunnel. Their only “innovation” in terms of any cost savings is to dig smaller tunnels - which we already knew could be done (tunnel cost grows with diameter), and which we don’t do for good reasons (capacity, emergency egress).
The branding and marketing exercise was excellent though.
Don't forget that Musk also founded Open AI (ChatGPT).
Lol this is why you aren't a VC. Even if every single Musk venture failed other than SpaceX, the investments would have paid off wildly well. You aim for the tails not the median.
In a way, its kind of cool to see how robber barons work in real time in our generation. Its also insanely depressing as they will systematically enshittify and extract as much wealth from society as is possible.
I don't actually think the Robber Barons in the 1920s had people going out of their way to defend them and insist they had special knowledge.
The New Deal happened with massive popular support because people did not like the Barons, and wanted to stop them and actually have a life worth living.
It only took like 30 years of suffering.
The Robber Barons weren't in the 1920s; that refers to industrial age monopolists (e.g. rail/oil), and culminated in the Sherman Antitrust (i.e. 1800s).
Broadly, your point is still valid, though. Just a mild inaccuracy between the Gilded Age and the roaring 20s.
Yeah, but there are enough people to buy the hype.
The guy also was super excited to go rap3 some k1ds in the island. Gladly he also failed at this.
Not sure if this is stock manipulation trying to push the bubble a little further in a way that doesn't require showing something of substance immediately, or if Elon's having another manic episode after doing too many or not enough drugs, but who didn't see this coming?
I'm sure next week he'll have SpaceX be bought by The Boring Company, sell that to Tesla, then rename all the companies as "X".
Also, whatever happened to his plan to turn twitter into a financial services company?
> "The only logical solution therefore is to transport these resource-intensive efforts to a location with vast power and space. I mean, space is called 'space' for a reason. [crying laughing emoji]"
This is all the reasoning provided. It is quite sad how a company I admired so much has become embroiled in financial doohickery.
It’s not SpaceX’s fault. It’s still a company to admire, it’s just that nobody appears to be able to stop Musk.
I wonder why SpaceX investors aren’t revolting.
I wonder why the SpaceX top management is going along with this. It's clearly not in their long term interest to do so.
> Starship will be capable of landing massive amounts of cargo on the Moon […] to establish a permanent presence and take advantage of lunar resources to manufacture satellites and deploy them further into space.
The trouble with strip mining the moon is that it is a pristine international geological park where one side is permanently visible from Earth*. In terms of park visits it’s been seen by pretty much every human that ever existed. Take that, Yosemite. The far side will be banned from exploitation to maintain its unique park status as being almost completely radio silent.
Perhaps the mining will take place behind the ridge line of limbward mountains: technically on the near side but without being visible. Going underground feels like a bit of a stretch.
On the far side, how far does one have to be from the anti-Earth point before one can fire up the WiFi without pissing off the space telescopes?
Who will even regulate this stuff? Do we extend the Antarctic treaty for whole-lunar purposes?
*Worst case, a 5km wide strip mine is 10 pixels on a DSLR photo, but that’s still too much for some.
If you had told me 4 years ago that Twitter would be merged into SpaceX I would have called you crazy. Yet here we are..
Is xAI Twitter? I thought they were separate companies, but honestly I don’t know anymore.
Yes X was merged into xAI last year I believe.
That’s wild. So it’s true that Twitter is merging with SpaceX?
That's what the headline is, yep. Wild times indeed.
Yep.. "all stock deal" last Spring. https://www.paddle.com/news/industry/elon-musk-xai-acquires-...
Elon Musk’s AI startup, xAI (x.ai = grok), officially acquired X Corp. (the parent company of the social media platform X or x.com, formerly Twitter/x.com) in an all-stock deal. Both now operate under a unified holding entity, frequently referred to in corporate filings as X.AI Holdings Corp. (or simply xAI). Now SpaceX has moved to acquire or merge with xAI. This effectively brings the social media data from x.com, the AI development of x.ai, and the satellite infrastructure of Starlink/SpaceX under one "super-conglomerate" roof.
the X super-conglomerate
I'm just waiting for the year 2350 when Walmart buys Weyland-Yutani.
> The basic math is that launching a million tons per year of satellites generating 100 kW of compute power per ton would add 100 gigawatts of AI compute capacity annually, with no ongoing operational or maintenance needs. Ultimately, there is a path to launching 1 TW/year from Earth.
> My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space.
This is so obviously false. For one thing, in what fantasy world would the ongoing operational and maintenance needs be 0?
You operate them like Microsoft's submerged data center project: you don't do maintenance, whatever fails fails. You start with enough redundancy in critical components like power and networking and accept that compute resources will slowly decrease as nodes fail
No operational needs is obviously ... simplified. You still need to manage downlink capacity, station keeping, collision avoidance, etc. But for a large constellation the per-satellite cost of that would be pretty small.
How do you make a small fortune? Start with a big one.
The thing being called obvious here is that the maintenance you have to do on earth is vastly cheaper than the overspeccing you need to do in space (otherwise we would overspec on earth). That's before even considering the harsh radiation environment and the incredible cost to put even a single pound into low earth orbit.
If you think the primary source of electricity is solar (which clearly Musk does), then space increases the amount of compute per solar cell by ~5x, and eliminates the relatively large battery required for 24/7 operation. The thermal radiators and radiation effects are manageable.
The basic idea of putting compute in space to avoid inefficient power beaming goes back to NASA in the 60s, but the problem was always the high cost to orbit. Clearly Musk expects Starship will change that.
My dude, ISS has 200 KW of peak power.
NVIDIA H200 is 0.7 KW per chip.
To have 100K of GPUs you need 500 ISSs.
ISS cooling is 16KW dissipation. So like 16 H200. Now imagine you want to cool 100k instead of 16.
And all this before we talk about radiation, connectivity (good luck with 100gbps rack-to-rack we have on earth), and what have you.
—
Sometimes I think all this space datacenters talk is just a PR to hush those sad folks that happen to live near the (future) datacenter: “don’t worry, it’s temporary”
https://www.nytimes.com/2025/10/20/technology/ai-data-center...
The ISS is in the middle of rolling out upgrades to their panels so it’s not a great comparison. https://en.wikipedia.org/wiki/Roll_Out_Solar_Array
> ROSA is 20 percent lighter (with a mass of 325 kg (717 lb))[3] and one-fourth the volume of rigid panel arrays with the same performance.
And that’s not the current cutting edge in solar panels either. A company can take more risks with technology choices and iterate faster (get current state-of-the-art solar to be usable in space).
The bet they’re making is on their own engineering progress, like they did with rockets, not on sticking together pieces used on the ISS today.
Now tell me how you heat dissipate all this. Not that there is a lot of air or water in space.
Not that you would want 500+ square meters just for cooling of 200KW
And, mind you, it won’t be a simple copper radiator
https://www.nasa.gov/wp-content/uploads/2021/02/473486main_i...
How much maintenance do you need? Lets say you have hardware whose useful lifespan due to obsolescence is 5 years, and in 4, the satellite will crash into the atmosphere anyways.
Let's say given component failure rates, you can expect for 20% of the GPUs to fail in that time. I'd say that's acceptable.
> How much maintenance do you need?
A lot. As someone that has been responsible for trainings with up to 10K GPUs, things fail all the time. By all the time I don't mean every few weeks, I mean daily. From disk failings, to GPU overheating, to infiniband optical connectors not being correctly fastened and disconnecting randomly, we have to send people to manually fix/debug things in the datacenter all the time.
If one GPU fails, you essentially lose the entire node (so 8 GPUs), so if your strategy is to just turn off whatever fails forever and not deal with it, it's gonna get very expensive very fast.
And thats in an environment where temperature is very well controlled and where you don't have to put your entire cluster through 4 Gs and insane vibrations during take off.
Radiation is a bitch. Especially at those nanometers and memory bandwidth.
And cooling. There is no cold water or air in space.
The idea here is that the economics of launch are changing with Starship such that the "incredible cost" and "overspeccing" of space will become much less relevant. There's a world where, because the cost per kg is so low, a data center satellite's compute payload is just the same hardware you'd put in a terrestrial rack, and the satellite bus itself is mass-produced to not-particularly-challenging specs. And they don't have to last 30 years, just 4-ish, when the computer is ready for retirement anyway.
Will that come to be? I'm skeptical, especially within the next several years. Starship would have to perform perfectly, and a lot of other assumptions hold, to make this make sense. But that's the idea.
If you ramp up the economies of scale to make those things - radiation protection and cost per pound - the calculus changes. It's supposed to synergize with Starship, and immediately take advantage of the reduced cost per pound.
If the cost per pound, power, regulatory burden, networking, and radiation shielding can be gamed out, as well as the thousand other technically difficult and probably expensive problems that can crop up, they have to sum to less than the effective cost of running that same datacenter here on earth. It's interesting that it doesn't play into Jevon's paradox the way it might otherwise - there's a reduction in power consumption planetside, if compute gets moved to space, but no equivalent expansion since the resource isn't transferable.
I think some sort of space junk recycling would be necessary, especially at the terawatt scale being proposed - at some point vaporizing a bunch of arbitrary high temperature chemistry in the upper atmosphere isn't likely to be conducive to human well-being. Copper and aluminum and gold and so on are also probably worth recovering over allowing to be vaporized. With that much infrastructure in space, you start looking at recycling, manufacturing, collection in order to do cost reductions, so maybe part of the intent is to push into off-planet manufacturing and resource logistics?
The whole thing's fascinating - if it works, that's a lot of compute. If it doesn't work, that's a lot of very expensive compute and shooting stars.
some people really gotta stop huffing VC fumes
Or, just saying, be critical of ideas and think them through, and take in what experts say about it, and determine for yourself what's up. If a bunch of people who usually seem to know what they're talking about think there's a legitimate shot at something you, as a fellow armchair analyst, think is completely impractical, it makes sense to go and see if maybe they know something you don't.
In this case, it's all about Starship ramping up to such a scale that the cost per pound to orbit drops sufficiently for everything else to make sense - from the people who think the numbers can work, that means somewhere between $20 and $80 per pound, currently at $1300-1400 per pound with Falcon 9. Starship at scale would have to enable at least 2 full orders of magnitude decrease in price to make space compute viable.
If Starship realistically gets into the $90/lb or lower range, space compute makes sense; things like shielding and the rest become pragmatic engineering problems that can be solved. If the cost goes above $100 or so, it doesn't matter how the rest of the considerations play out, you're launching at a loss. That still might warrant government, military, and research applications for space based datacenters, especially in developing the practical engineering, but Starship needs to work, and there needs to be a ton of them for the datacenter-in-space idea to work out.
Or, just saying, we should eat babies because they are abundant and full of healthy nutrition for adult humans. [1]
Just because an idea has some factors in its favor (Space-based datacenter: 100% uptime solar, no permitting problems [2]) doesn't mean it isn't ridiculous on its face. We're in an AI bubble, with silly money flowing like crazy and looking for something, anything to invest it. That, and circular investments to keep the bubble going. Unfortunately this gives validation to stupid ideas, it's one of the hallmarks of bubbles. We've seen this before.
The only things that space-based anything have advantages on are long-distance communication and observation, neither of which datacenters benefit from.
The simple fact is that anything that can be done in a space-based datacenter can be done cheaper on Earth.
[1] https://en.wikipedia.org/wiki/A_Modest_Proposal for the obtuse
[2] until people start having qualms about the atmospheric impact of all those extra launches and orbital debris
Note how Musk cleverly doesn't claim that not doing maintenance drives down costs.
Nothing in there is a lie, but any substance is at best implied. Yes, 1,000,000 tons/year * 100kW/ton is 100GW. Yes, there would be no maintenance and negligible operational cost. Yes, there is some path to launching 1TW/year (whether that path is realistic isn't mentioned, neither what a realistic timeline would be). And then without providing any rationale Elon states his estimate that the cheapest way to do AI compute will be in space in a couple years. Elon is famously bad at estimating, so we can also assume that this is his honest belief. That makes a chain of obviously true statements (or close to true, in the case of operating costs), but none of them actually tell us that this will be cheap or economically attractive. And all of them are complete non-sequiturs.
Elon might have a scoop on getting things to orbit cheaper than everyone else.
Unless I missed something the Microsoft underwater data center was basically a publicity stunt.
Anyone who thinks it makes sense to blast data centers into space has never seen how big and heavy they are, or thought about their immense power consumption, much less the challenge of radiating away that much waste heat into space.
Radiation is an even bigger problem, especially in the polar orbits they are talking about.
It’s only a problem if you get the machines up there! Which I’d argue is economically unviable to boot.
I don't think it was a stunt. It was an experiment.
I think passive cooling (running hot) reduced some of the advantages of undersea compute.
I was listening to a Darknet Diaries episode where Maxie Reynolds seems to make it work: https://subseacloud.com/ I don't know how profitable they are, and I doubt this is scalable enough, but it can work as a business.
Ironically a benefit of underwater datacenters would be reduced cosmic rays. Not so great in orbit, I imagine!
What about a data centre only running SQLite?
Well the thing is that it seemed to have been successful beyond all expectations despite being that? They had fewer failures due to the controlled atmosphere, great cooling that took no extra power, and low latency due to being close to offshore backbones. And I presume you don't really need to pay for the land you're using cause it's not really on land. Can one buy water?
Space is pretty ridicolous, but underwater might genuinely be a good fit in certain areas.
If it was successful beyond all expectations, why aren't we seeing more?
Hot saltwater is the worst substance on earth, excepting, maybe, hydrofluoric acid. You really don't want to cool things with ocean water over an extended period of time. And filtering/purifying it takes vast amounts of power (e.g. reverse osmosis).
My 4 Cylinder Diesel Volvo Penta is cooled by sea water. There is an elbow that may have to be replaced every few years,
I wonder why they did not start with a freshwater body.
I thought they had an issue with stuff growing on the cooling grates. Life likes to find warm water.
An 8 GPU B200 cluster goes for about $500k right now. You'd need to put thousands of those into space to mimic a ground-based data center. And the launch costs are best case around 10x the cost of the cluster itself.
Letting them burn up in the atmosphere every time there's an issue does not sound sustainable.
A Falcon Heavy takes about 63 tons to LEO, at a cost of about $1,500 per kg. A server with 4x H200s and some RAM and CPU costs about $200k, and weighs about 60kg, with all the cooling gear and thick metal. As is, it would cost $90k to get to LEO, half of the cost of the hardware itself.
I suppose that an orbit-ready server is going to cost more, and weigh less.
The water that serves as the coolant will weigh a lot though, but it can double as a radiation shield, and partly as reaction mass for orbital correction and deorbiting.
Just so we can agree on numbers for the napkin math - an 8x H200 weighs 130 kg:
https://www.nvidia.com/en-eu/data-center/dgx-h200/?utm_sourc...
Power draw is max 10.2 kW but average draw would be 60-70% of that. let's call it 6kW.
It is possible to obtain orbits that get 24/7 sunlight - but that is not simple. And my understanding is it's more expensive to maintain those orbits than it would be to have stored battery power for shadow periods.
Average blackout period is 30-45 minutes. So you'd need at least 6 kWh of storage to avoid draining the batteries to 0. But battery degradation is a thing. So 6 kWh is probably the absolute floor. That's in the range of 50-70 kg for off-the-shelf batteries.
You'd need at least double the solar panel capacity of the battery capacity, because solar panels degrade over time and will need to charge the batteries in addition to powering the gpu's. 12 kW solar panels would be the absolute floor. A panel system of that size is 600-800 kg.
These are conservative estimates I think. And I haven't factored in the weight of radiators, heat and radiation shielding, thermal loops, or anything else that a cluster in space might need. And the weight is already over 785 kg.
Using the $1,500 per kg, we're approaching $1.2 million.
Again, this is a conservative estimate and without accounting for most of the weight (radiators) because I'm too lazy to finish the napkin math.
I think we're on the same page. Lifting the actual computing devices would be not that expensive, compared to lifting a lot of other related mass, principally the cooling systems, and the solar panels.
The solar panels used in space are really lightweight, about 2 kg / m² [1], it's like ten times lighter weight than terrestrial panels. Still they need load-bearing scaffolding, and electrical conductors to actually collect the hundreds of kilowatts.
Water can't be made as lightweight though.
[1]: https://space.stackexchange.com/a/30238
Are launch costs really 10x!? Could I get a source for that?
In the back on my head this all seemed astronomically far-fetched, but 5.5 million to get 8 GPUs in space... wild. That isn't even a single TB of VRAM.
Are you maybe factoring in the cost to powering them in space in that 5 million?
I guess he adds the weight of all the hardware to make the whole thing work.
You also need square kms of radiators to cool 100MW
Playing devil's advocate, when a GPU dies you don't typically fix it, right? You just replace it.
What if you could keep them in space long enough that by the time they burn up in the atmosphere, there are newer and better GPUs anyway?
Still doesn't seem sustainable to me given launch costs and stuff (hence devil's advocate), but I can sort of see the case if I squint?
Let me rip my bong real quick..
What if you had a fleet of Optimus robots up there who would actually operate a TSMC in space and they would maintain the data centers in space?
Hold on let me enter a K hole…
What if we just did things?
YC founder spotted
You mean you operate them like Microsoft's failed submerged data center project [1]. When pointing at validating past examples you are generally supposed to point at successes.
[1] https://en.wikipedia.org/wiki/Project_Natick
The opposite of down is up, so it wouldn't be completely illogical.
Did we read the same Wikipedia page? It doesn't say the word "failed" anywhere on it.
> By 2024, Project Natick had been inactive for several years, though it was referenced in media as though it was ongoing. That year, Microsoft confirmed that the project was inactive and that it had no servers underwater.
I wouldn't exactly call this a success, for that matter.
But if we’re going down that line of thinking then it’s a poor comparison. I could open a data centre on the ground and use the same principle of zero maintenance, and it would be way cheaper and way more powerful.
But you could just run a “zero maintenance” data center on Earth and not pay to blast it into orbit.
How many submerged datacenters is Microsoft operating?
None, https://www.datacenterdynamics.com/en/news/microsoft-confirm...
Zero.
You also have to get rid of waste heat :P
This will totally work since we have an unlimited amount of rare earth elements we can just ship off into space never to see again. Infinite raw materials + infinite power equals infinite AI!!!
The financial system is already freaking out about depreciation on land based data centers. I don't think it could survive what you're talking about.
Accountants love this
Being under the ocean in a metal box you don't get too many micro-meteors or cosmic rays though.
...and costs pennies compared to putting anything up there, so it can even enjoy those cosmic goodies.
Just like you don’t get much water in space.
My understanding was that access to very large body of cold water was a core feature for the project. The water was to be used for cooling relatively efficiently or cheaply.
As soon as a statement contains a timeframe estimate by Musk you know to disregard it entirely.
The thing is: at the end of the day, SpaceX takes the "impossible" and makes it "late".
People are going to Tory Bruno the space datacenters until one day their Claude agent swarm's gonna run in space and they'll be wondering "how did we get here"?
The thing is: at the end of the day, making absolute statements about the inevitability of future success is a fool’s errand.
Musk has a documented history of failing to deliver on promises, timescale or no. So it’s best to engage in some actual critical thinking about the claims he is making.
napkin math says sq kms of radiators to cool 100MW, it's just patently ridiculous
What if they use heat pumps to raise the temperature? Heat rejection is proportional to T^4.
Or takes the impossible and puts a half baked version of it behind a $99/ month paywall.
Pay wallet? Starlink was never gonna be free.
There's clearly rhetorical hyperbole happening there. But assuming that thermal rejection is good in space, & launch costs continue falling, as earth-based data centers become power/grid-constrained, there is a viable path for space power gen.
The craziest part of those statements is "100 kW per ton." IDK what math he is doing there or future assumptions, but today we can't even sniff at 10 kW per ton. iROSA [1] on the ISS is about 0.150 kW per ton.
[1]https://en.wikipedia.org/wiki/Roll_Out_Solar_Array
edit: iROSA = 33 kW per ton, thanks friends
Not to be an Elon defender, but can you back up your 0.15/ton? My own searching puts ROSA orders of magnitude higher. Each array is 600kg (0.6t) and puts out 20kw (https://space.skyrocket.de/doc_sdat/irosa-1.htm) which makes 20/0.6 = 33.333 kw/ton
You're right, my fault. I made an math booboo somewhere. Your calc seems right
Hey all good. My advice, not that you asked for it, is to put the math in the comment. Even as a footnote. I've found myself backtracking a lot of math comments after I stare at it in the text box for a few seconds.
The company lists their ISS solar panels as 28 kW for 331 kg, which comes pretty near to 100 W/kg.
Company website:
https://rdw.com/wp-content/uploads/2023/06/redwire-roll-out-...
And their Opal configuration beats the metric: 5.3 kW for 42.7 kg.
Okay so that works out to 124 kW/ton for the opal config.
Maybe I'm just out of the loop, but is solar substantially more efficient in space? I assume the satellites won't orbit in a way that follows the sun. And presumably the arrays of panels they can attach to a satellite don't exceed the size of the panels you could slap on and around a data center (at least without being insanely expensive).
Yeah the main benefits are:
1. solar is very efficient at generating energy, no moving parts, simple physics etc.
2. in space you don't deal with weather or daylight cycle, you can just point your panels at the sun and generate very stable energy, no batteries required
3. environmental factors are simpler, no earthquakes, security, weather. Main problem here is radiation
In theory its a very elegant way to convert energy to compute.
2 is wrong. At a Lagrange point you can do this. Not in low earth orbit - in LEO sunset is every 60 minutes or so, and you spend the next 60 minutes in darkness.
Satellites are heavily reliant on either batteries or being robust to reboots, because they actually do not get stable power - it's much more dynamic (just more predictable too since no weather).
Interesting.
According to this other source https://www.satellitetoday.com/connectivity/2026/02/02/space...
the filing mentions this
> these satellites would operate between 500 km and 2,000 km altitude and 30 degrees and Sun-Synchronous Orbit inclinations (SSO)
> But assuming that thermal rejection is good in space
Don't assume this. Why would you assume this?
Just put a slightly larger solar array on the same equipment on earth?
> put a slightly larger solar array on the same equipment on earth?
Land and permitting. I’m not saying the math works. Just that there are envelopes for it to.
There is practically infinite land in which to build a datacenter.
> There is practically infinite land in which to build a datacenter
This is absolutely not true. I’ve worked on some of this stuff. Permitting costs months, which in dollar terms pays for launch costs ten-fold.
The math literally works.
The US mandates by law that we grow a fuck ton of corn to mix 10% ethanol into gasoline.
If you replaced just those cornfields with solar/wind, they would power the entire USA and a 100% electric vehicle fleet. That includes the fact that they are in the corn belt with less than ideal sun conditions.
We aren’t even talking about any farmland that produces actual food or necessary goods, just ethanol as a farm subsidy program.
The US is already horrendously bad at land use. There’s plenty of land. There’s plenty of ability to build more grid capacity.
Hello fello Technology Connections watcher?
You know it!
Solar in space is a very different energy source in terms of required infrastructure. You don't need batteries, the efficiency is much higher, cooling scales with surface area (radiative cooling doesn't work as well through an atmosphere vs. vacuum), no weather/day cycles. Its a very elegant idea if someone can get it working.
Only if you also disregard all the negatives.
The panels suffer radiation damage they don't suffer on Earth. If this is e.g. the same altitude orbits as Starlink, then the satellites they're attached to burn up after around tenth of their ground-rated lifetimes. If they're a little higher, then they're in the Van Allen belts and have a much higher radiation dose. If they're a lot higher, the energy cost to launch is way more.
If you could build any of this on the moon, that would be great; right now, I've heard of no detailed plans to do more with moon rock than use it as aggregate for something else, which means everyone is about as far from making either a PV or compute factory out of moon rock as the residents of North Sentinel Island are.
OK, perhaps that's a little unfair, we do actually know what the moon is made of and they don't, but it's a really big research project just to figure out how to make anything there right now, let alone making a factory that could make them cost-competitive with launching from Earth despite the huge cost of launching from Earth.
> The panels suffer radiation damage they don't suffer on Earth.
I don't think this is true, Starlink satellites have an orbital lifetime of 5-7 years, and GPUs themselves are much more sensitive than solar panels for rad damage. I'd guess the limiting factor is GPU lifetime, so as long as your energy savings outpace the slightly faster gpu depreciation (maybe from 5 -> 3 years) plus cost of launch, it would be economical.
I've said this elsewhere, but based on my envelope math, the cost of launch is the main bottleneck and I think considerably more difficult to solve than any of the other negatives. Even shielding from radiation is a weight issue. Unfortunately all the comments here on HN are focused on the wrong, irrelevant issues like talking about convection in space.
Sun-synchronous orbit means there's no nightime for satellites in that orbit.
"There's clearly rhetorical hyperbole happening there" in a business paper is called lying
Name a unicorn whose early round pitch decks are 100% free of wishful thinking
Elon is a pathological liar and it’s crazy that he still gets sanewashed after all he’s done. It’s insanity that he hasn’t been kicked out of leading his companies, and it’s also insanity that he hasn’t been prosecuted by the SEC.
You’ve spent too much life force trying to even understand the liar’s fake logic.
Let’s start right here: there is no such thing as becoming power/grid constrained on earth. If you replaced just the cornfields that the United States uses just to grow corn for ethanol in gasoline just in the corn belt, you could power the entire country with solar+batteries+wind. Easily, and cheaply.
If you don’t even believe that solar+batteries are cheap (they are), fine, choose your choice of power plant. Nuclear works fine.
The truth is, xAI combining with SpaceX is almost certainly corrupt financial engineering. SpaceX as a government contractor and that means Elon’s pal Trump can now siphon money into xAI via the federal government.
>SpaceX as a government contractor and that means Elon’s pal Trump can now siphon money into xAI via the federal government.
I wonder how much faith Musk has that the US will never again have a president and/or Congress willing to torpedo such an incestuous deal.
This is par for the course for an Elon-associated endeavor but it's been leaking out into the broader tech sector; make ludicrous claims and promises and somehow investors just throw money at you. FSD has been around the corner for over a decade, martian colonization will be here by the end of the decade for the past 20 years and General SuperAI will be here in a few years for the past few years.
Currently, just a cursory google search shows $1500-3000 per kilogram to put something into low earth orbit. Lets take the low bound because of efficiencies of scale. So $1500.
A million tons will cost $1500x1000x1000000= 1,500,000,000,000. That is one and a half TRILLION dollars per year. That is only the lift costs, it does not take into account the cost of manufacturing the actual space data centers. Who is going to pay this?
That's the price before Starship which would be the prerequisite for this whole project.
Yes, and as we know Starship will be doing regular commercial launches starting in 2020, maybe 2021.
We're getting close to having the time for Starship's delays to be the same as the actual time for the Saturn 5 to go from plans to manned launches (Jan 1962-Dec 1968).
Are you trying to say it'll be delayed or that it'll never work?
One is obviously true, and the other is very likely false.
It’s hard to estimate what Starship’s actual costs will be when it isn’t fully operational. I am finding estimates of $100 to $200 per kilogram and even as low as $10 per kilogram.
Let’s say the costs in 5 years do get as low as $15 per kilogram or about 2 orders of magnitude improvement in launch prices. That means a 200-ton payload Starship would cost $3,000 to launch.
Do you honestly believe that? The world’s largest rocket cost a total of $3,000 to launch?
> Let’s say the costs in 5 years do get as low as $15 per kilogram or about 2 orders of magnitude improvement in launch prices. That means a 200-ton payload Starship would cost $3,000 to launch. Do you honestly believe that? The world’s largest rocket cost a total of $3,000 to launch?
You have missed three zeroes in this calculation ;)
15 per kg for a 200-ton payload is about 3 million$. That seems achievable, given that propellant costs are about 1-1.5 million.
Ah yeah. 200 tons is 200,000 kilograms. Definitely way off there. That is an incredulous number.
"it'll never work" is quite black and white while "failure" is a lot more of a grey area. Will it actually launch? Sure, we've seen it. Will it actually hit the reliability as sold? Will it have as fast of turnaround time to reach launch timing goals? Can it actually launch as much payload as promised? Will the economics actually shake out as intended?
Did the Cybertruck "never work"? Obviously not, they're on the streets. Was it a <$40k truck with >250mi range? No.
Did FSD "never work"? Obviously not, tons of people drive many, many miles without touching the wheel. Does Tesla feel confident in it enough to not require safety operators to follow it on robotaxi trips? No. Does Tesla trust it enough to operate in the Las Vegas Loop? No. Has Tesla managed to get any state to allow it to operate truly autonomously? No.
Look, I hope Starship does work as advertised. Its cool stuff. But I don't see it as a given that it will. And given by the track record of the guy who promised it, it gives even less confidence. I'm sad there's less competition in this space. We have so many billionaires out there and yet so few out there actually willing push envelopes.
One reliable method of pushing envelopes, attracting investment and hiring smart people is to get excited about unrealistic timelines.
The best case is you meed the unrealistic timeline, the average case outcome is you solve the problem but it is delayed several years. And the worst case is it fails and investors lose some money.
If you try to hire people but your message is: we want to reduce the cost of access to space by 20% in thirty years, you are going to get approximately zero competent engineers, and a whole lot of coasters.
And no investors, so you'll be dependent on the government anyway. Depending on the government is great until people you do not agree with or are generally anti science, are in power. I assume this part should not need an example nowadays?
> One reliable method of pushing envelopes, attracting investment and hiring smart people is to get excited about unrealistic timelines.
Its also a good way to shred morale and investor confidence when you're a decade past your timelines or continue to fail on actually delivering on past promises.
You'd think so, but if you bet on this guy not being able to get investors you'll end up being wrong.
It doesn't make sense (neither does Tesla's valuation, for example), but it is what it is.
Both Spacex and Xai have investors lining up.
That launch cost is remarkably cheap to someone that's handled a $1.5million dollar 5U server filled with GPUs and RAM that weighs under 100kg.
Obviously the solar and cooling for the above would both weigh and cost a ton but... It's feels surprisingly close to being within an order of magnitude of current costs when you ballpark it?
Like i don't think it's actually viable, it's just a little shocking that the idea isn't as far out of line as i expected.
AI and space based economy
It's always 2-3 years with this guy
Conveniently, about the amount of time it takes for the average person to forget and/or rematerialize in a new parallel dimension
The average person is not aware or does not care about Elon Musk’s claims and whether or not they come true.
You’re just not going at the speed of light as this guy’s brain is, time dilation is a thing
> in what fantasy world would the ongoing operational and maintenance needs be 0?
Well, if you can't get there, you can't do maintenance, so there is zero maintenance :)
Satellites have large operational costs. Satellite fleets even more so.
I remember reading somewhere that satellites are extra expensive for 2 reasons:
- launch costs are so high that doing exotic bespoke engineering might be worth it if it can shave off a few pounds
- once again because launches are expensive and rare, you cannot afford to make mistakes, so everything has to work perfectly
If you are willing to launch to lower orbits, and your launch vehicle is cheap, you are building in bulk, then you can compromise on engineering and accept a few broken sats
Undergrads afaik even high schoolers have built cubesats out of aluminum extrusions, hobbyist solar panels, and a tape measure as an antenna. These things probably dont do that much, but they are up there and they do work.
They are also expensive because there are unique challenges to making reliable spacecraft. E.g. cosmic rays and microgravity absolutely wreck electronics. Those undergrad cube sats are lucky to last more than a few months in the relative calm of low inclination, low altitude earth orbit. They would die on their first pass in a sun synchronous polar orbit.
Pet satellites or cattle satellites?
Kessler satellites
The ISS’s solar arrays each weigh a metric ton and generate 35 KW a piece[0], and that’s just for the power collection.
They’d need incredible leaps in efficiency for an orbiting ton collecting and performing 100 KW of compute.
[0] https://en.wikipedia.org/wiki/Electrical_system_of_the_Inter...
The famous Musk timeline. "By next year, 2 year tops".
By his original timeline we should have landed on Mars… 4 years ago.
https://www.planetary.org/articles/20170929-spacex-updated-c...
and "Full Self-Driving" was a solved problem in 2016.
Ooh, happy 10th anniversary, FSD?
Here is my main question: Musk is on record as being concerned about runaway "evil AI." I used to write that off as sci-fi thinking. For one thing, just unplug it.
So, let's accept that Musk's concern of evil runaway AI is a real problem. In that case, is there anything more concerning than a distributed solar powered orbital platform for AI inference?
Elon Musk appears to be his own nemesis.
He just says stuff to convince people of things that benefit him. Internal consistency was never the plan.
My point is not to make fun of him, but to help avoid the destruction of humanity via an HN comment. No joke.
This is starting to get really serious.
Aside from anything about Elon Musk, here’s an interesting video response to the “just unplug it” argument on the Compuerphile channel: https://youtu.be/3TYT1QfdfsM
Ha, I figured that might be the video prior to clicking it. I am a long time fan.
Agreed, when I wrote "just unplug it," this counterargument was present in my mind, but nobody likes a wall of text.
However, my original point was that a distributed solar powered orbital inference platform is even worse! Think about how hard it would be to practically take out Starlink... it's really hard.
Now.. >1M nodes of a neural net in the sky? Why would someone who lives as a god, the richest man in the world, the only person capable of doing this thanks to his control of SpaceX... do the literal worst thing possible?
That'd easily take a few LEO detonated fragmentation bombs to trigger a cascading LEO shrapnel field.
It's a lot harder than taking out some terrestrial power lines.
Sure, it'd take obital launch capabilities to lift ... how many bags of metal scrap and explosives?
tone: I don't really understand orbital mechanics, but I do understand geopolitics a bit.
1. China is very concerned about Starlink-like constellations. They want their own, but mostly they want to be able to destroy competitors. That is really hard.
2. Many countries have single ASAT capabilities. Where one projectile can hit one satellite. However, this is basically shoot a bullet, with a bullet, on different trajectories.
3. > Sure, it'd take orbital launch capabilities to lift ... how many bags of metal scrap and explosives?
If I understand orbital mechanics... those clouds of chaff would need to oppose the same orbit, otherwise it is a gentle approach. In the non-aligned orbit, it's another bullet hitting a bullet scenarios as in 2, but with a birdshot shotgun.
My entire point is that constellations in LEO take hundreds of Falcon 9's worth of mass to orbit and delta-v to destroy them, as in-orbit grenades which approach gently. This IS REALLY HARD, as far as mass to orbit, all at once! If you blow up some group of Starlink, that chaff cloud will just keep in orbit on the same axis. It will not keep blowing up other Starlinks.
The gentle grenade approach was possibly tested by the CCP here:
https://news.ycombinator.com/item?id=46820992
> tone: I don't really understand orbital mechanics, but I do understand geopolitics a bit.
Thanks for the clarification, I guess that explains this (from you):
> Think about how hard it would be to practically take out Starlink.
and this:
> My entire point is that constellations in GEO
which you've now corrected.
Moving on:
> My entire point is that constellations in LEO take hundreds of Falcon 9's worth of mass to orbit and delta-v to destroy them, as in-orbit grenades which approach gently. This IS REALLY HARD
So let's not do that .. how hard is it to render the entire LEO zone a shit show with contra wise clouds of frag that cause cascading failures?
Forget the geopolitics of China et al. .. LEO launch capabilities are spreading about the globe, it's not just major world powers that pose a threat here.
Ok... so, let's reset, please. I bet that we have very similar intentions, and yet on internet forums, we have perfected the art of users speaking past each other.
Just to get on the same page here. My arugument is that prior to Elon Musk, the only human capable of launching >1M distributed solar powered inference nodes, if one accepts runaway AGI/ASI as a threat... prior to that we had a few hundred terrestrial AI inference mega-data centers. Most of them had easily disrupted power supplies by one dude with a Sawzall.
Now, we are moving to a paradigm where the power supply is the sun, the orbital plane gives the nodes power 24/7, and the dude with the Sawzall needs to buy >10,000x (not sure of the the multiple here) the Sawzalls, and also give them escape velocity.
Can we not agree that this is a much more difficult problem to "just unplug it," than it was when the potentially troublesome inference was terrestrial?
There are many people in this world who, if asked, would regard taking out a LEO constellation as an interesting challenge.
My up thread commentary was not meant as real snark at all. I was attempting to be genuine.
However, I think it did accomplish my goal. I bet that we could now have a beer/tea, and laugh together.
If you are ever near Wroclaw, Prague, Leipzig/Dresden, or Seattle, please email my username at the the big G. I would happily meet you at the nearest lovely hotel bar. HN mini meetup. I can only imagine the stories that we might exchange.
:-)
Look, I'm Australian, I enjoy a bit of banter. I stripped the personal info from my comment above; I was happy to share with you, reluctant to leave it as was.
I was a frequent Toronto visitor, for the TSX, back when we ran a minerals intelligence service before passing that onto Standard&Poor.
You're on the list, however my movements are constrained for now, my father's a feriously active nonagenarian which is keeping me with one foot nailed to the ground here for now.
Cheers to you and your father.
Also, thank you for the reminder that I need to get my ass back to Seattle to be with remaining parent, while I still can. I have been a jackass about that.
What, creating a huge patchwork of self sufficient AIs, forming their own sky based net, seems bad to you, considering the whole torment nexus/Sky Net connotations? It's not like he's planning to attach it to his giant humanoid robot program. Oh. Ohhhhh. Oh no.
> in what fantasy world
It is already more expensive to performance maintenance on SOCs than it is to replace them. Remember, these machines are not for serving a database, there are practically no storage needs (and storage is the component that fails most often.)
Given that, the main challenge is cooling, I assume that will be figured out before yeeting 100 billion $ of computers into space. Plenty of smart people work at these companies.
Do smart people work at Boring Company? Do smart people work on FSD at Tesla? What about the HyperLoop? It is possible for smart people to make technical achievements without the overall project being particularly successful.
You're right.
I meant it specifically for figuring out cooling computers in space.
I am pretty sure this is going to be a solvable problem if this is the bottleneck to achieve data centers in space, given that newer chips are much more tolerant to high temperatures.
https://www.marketplace.org/story/2026/01/07/new-ai-chips-wi...
Any estimate by Elon musk, you need to add or substract a zero to/from the end. Here, I'll fix it for you.
> The basic math is that launching a 100,000 tons per year of satellites generating 10 kW of compute power per ton would add 1 gigawatt of AI compute capacity annually, with no ongoing operational or maintenance needs. Ultimately, there is a path to launching 0.01 TW/year from Earth. > My estimate is that within 20 to 30 years, the lowest cost way to generate AI compute will be in space.
>This is so obviously false.
One of the biggest but most pointless questions I have about our current moment in history is whether the people in power actually believe the stuff they say or are lying. Ultimately I don't think the answer really matters, their actions are their actions, but there is just so much that is said by people like Musk that strains credulity to the point that it indicates either they're total idiots or they think the rest of us are total idiots and I'm genuinely curious which of those is more true.
We’re at a point where propaganda is so much more powerful than reality that the people in power literally can’t tell the difference. When your source of ethics is the stock price, little details like physical impossibility stop seeming relevant.
You put it so succinctly and perfectly that I'll have to favorite your comment. Totally agree. The physical world has become little more than noise for people like Musk. I wonder whether the correction will be a slow market dip, a full collapse, or somehow whether he makes it out like a bandit. Baudrillard is, once again, uncomfortably accurate in his diagnosis.
But more importantly, there is no heat dissipation in space. There’s no atmosphere to cool you, no water you can put heat into. Just an empty void. You can radiate a little, but the sun alone is enough to cook you, without you having a rack of GPUs inside your satellite.
It’s completely delusional to think you could operate a data centre in a void with nowhere to put the heat.
Do you honestly believe that nobody involved has ever considered that?
Apparently not, otherwise this silly idea wouldn't exist.
Naysayers probably get fired fast.
Launching alone consumes about 75-150kWh per tonne of energy for fuels only (as per ChatGPT).
Planned lifespan of Starlink satellites is 5years.
A million tons a year would be over 18 Starship launches per day.
Never mind operational and maintenance costs. In what fantasy world is it cheaper to put a computer in orbit than in a building on the ground? I don't care how reusable and maintenance-free Starship gets, there's no way even absurdly cheap launches are cheaper than a building.
The whole thing makes no sense. What's the advantage of putting AI compute in space? What's even one advantage? There are none. Cooling is harder. Power is harder. Radiation is worse. Maintenance is impossible.
The only reason you'd ever put anything in orbit, aside from rare cases where you need zero-gee, is because you need it to be high up for some reason. Maybe you need it to be above the atmosphere (telescopes), or maybe you need a wide view of the earth (communications satellites), but it's all about the position, and you put up with a lot of downsides for it.
I feel like either I'm taking crazy pills, or all these people talking about AI in space are taking crazy pills. And I don't think it's me.
The most generous interpretation is that the "AI in space" nonsense is a cover for putting limited AI in space for StarShield (military version of StarLink), which is essentially the "Golden Dome".
It might be possible to scam the Pentagon with some talk about AI and killer satellites that take down ICBMs.
Why would they need a cover, though? They can just say “we’re putting AI in space so we can shoot down missiles” and that would be fine.
> It might be possible to scam the Pentagon with some talk about AI and killer satellites that take down ICBMs.
Honestly that story sounds right up Pete Hegseth's alley.
You lost me on million tons.
They'll just be decommissioned and burn up in space. nVidia will make space-grade GPUs on a 2-3 year cycle.
You do realize that “space-grade” involves process changes that intrinsically incur orders of magnitude efficiency losses? Larger process sizes, worse performing materials. It’s not just a design thing you can throw money at.
Pre-requisites:
Ketamine
They don't need to be space grade, consumer hardware will do just fine.
For AI a random bit flip doesn't matter much.
Only if that bitflip happens somewhere in your actual data, vs. some GPU pipeline register that then locks up the entire system until a power cycle. Or causes a wrong address to be fetched. Or causes other nasty silent errors. Or...
Try doing fault injection on a chip some time. You'll see it's significantly easier to cause a crash / reset / hang than to just flip data bits.
'rad-triggered bit flips don't matter with AI' is a lie spoken by people who have obviously never done any digital design in their life.
As long as they stay below Van Allen belts and deal with weaker magnetic shielding in sun synchronous orbit (high latitudes).
I would say they probably something a little beefier than consumer hardware and just deal with lots of failures and bit flips.
But cooling is a bigger issue probably?
Random bit flips might even improve output.
Single upset events in a modern GPU are not bitflips. They destroy the surrounding circuitry and usually disable the whole unit.
If that happens you disable that CUDA core. If you GPU is too damaged, you deorbit the satellite.
Yeas, and this will happen within weeks of launch with the orbits under consideration.
> For one thing, in what fantasy world would the ongoing operational and maintenance needs be 0?
Do you not understand how satellites work? They don't send repair people into space.
This has been a solved problem for decades before the AI gold rush assumed they have some new otherworldly knowledge to teach the rest of the world.
> Do you not understand how satellites work?
Not trying to be rude - but it's you who doesn't understand how satellites work.
The U.S. has 31 GPS satellites in orbit right now. The operational cost of running those is $2 million/day.
Not to mention the scale of these satellites would be on the order of 10x-100x the size of the ISS, which we do send people to perform maintenance.
I fly satellites. None of them have a zero operational cost. None. Even the most automated cost money to keep running.
The problem is not solved and the techniques they use to deal with it run directly contrary to maximizing compute, because that's not historically something they have remotely cared about.
The problem is solved by making satellites extremely expensive.
<elon venture> rescues failing <elon venture> here have some <unattainable goals> the shareholders love that shiz.
Seems like a great way to play games with moving money around. Come up with a "valuation" and then "acquire".
It worked for his Twitter co-investors. I guess they overlap enough with Xai investors for them to think it's more clever than it is a rip off.
I'm pretty amazed one can play a shell game for so long and so openly with the public.
We have trouble cooling data centers on earth, so instead we are going to put them in space where cooling is a thousand times harder?
I think the idea is that energy production is a bigger bottleneck and easier in space.
Not saying I agree / disagree, but this seems to be the general thesis of people supporting this idea.
That, and the lack of regulation.
Politics and finances aside, I wonder how "sending megatons of mass" into space is more ecological than building power plants needed for data centers on earth? Not only all the fuel that you'd need to burn, but also the fact that this material probably can't be recycled since it will burn on reentry.
Just a neat bit of financial engineering. You can tell because Elon picked SpaceX instead of Tesla – which would have actually made sense at some level (Optimus Robots + AI). But Tesla is public and so he'd need to follow laws and reporting requirements.
You can tell it's just financial engineering because in the entire press release xAI is only mentioned in the first two sentences. Everything after that is Elon talking about space data centers to distract from the actual topic. Which seems to be working
does he need spacex/xai to prop up tesla or the other way around?
Tesla is still very profitable, as is SpaceX I assume. Twitter/X has been a $44 billion dollar failure, and xAI is a vanity project so Musk can go around saying he is a player in the AI space. Investors in both X and xAI need to be bailed out, hence this announcement.
Tesla has a P/E in the hundreds and a ~0.3% market cap to profit ratio. In what world is this "very" profitable?
In the world where it makes $8-10B in profit on $90-95 billion in revenue every year. Whatever price investors choose to trade the stock at is irrelevant to those numbers.
It's actually down to $3.8B in profit now, and will be losing money within a year at the rate its been losing profitability.
2% net return on assets is garbage
The $44B Twitter/X buyout was not a failure. For example Fidelity has its $19M investment in the buyout - now xAI common shares - marked at $62M (up over 3X) as of 12/31/25. It was certainly valued even higher on 1/31/26 after xAI had an oversubcribed fund raise in January. All before this merger announcement.
The fact that it had to be successively bailed out by xAI (which itself was funded by Tesla) and now SpaceX shareholders is exactly what makes the acquisition a failure.
He spent other people's money (or maybe even imaginary money) he couldn't have used for himself (since selling off major stakes in your company is a big nono)
A "bailout" is when a company rescued from bankruptcy. Common equity holders take large losses or are wiped out. This did not happen here.
We also know the Twitter buyout debt was sold at near par before the merger with xAI which is inconsistent with being near bankruptcy.
> xAI had an oversubcribed fund raise in January
My understanding is that it was not oversubscribed and would not have closed without Tesla’s investment.
It was originally for $15B. They raised $20B of which $2B was from Tesla.
Your sources might be shady (Elektrek?).
I think its an effort to position SpaceX as an AI company in order to justify some ridiculous valuation at IPO.
I think it's more so that the upcoming new public shareholders of SpaceX bail out his X/xAI misadventure.
Do you genuinely not think that "Elon" (xAI) is player in the AI space?
You don't have to think they have the best models of course, but they are clearly a very significant, and some might argue, leading player in the AI race.
> and some might argue, leading player in the AI race
What is this argument exactly? What are they leading?
It is a real model, real datacenters, and deployed heavily on their social media platform.
That's the full stack? Only other player that vertically setup is facebook, google and microsoft.
xAI’s models are really not pioneering at all. They weren’t the first to do MoE. Not the first to do open weighting, not the first to have memory or multi-modal vision.
So no, I wouldn’t say Elon is a major player in the AI space. People use his models because they are cheap and are willing to undress people’s photos.
saying they aren't pioneering is very different than saying they aren't a major player in the space. There're only like 5-7 players with a foundational model that they can serve at scale. xAI is one of them
I suspect SpaceX will acquire Tesla at some point. It’s the most profitable of these companies. So basically SpaceX employees and shareholders are covering up for the failing Tesla business and the already-failed xAI business.
Let’s not forget, xAI is the parent of Twitter/X (the social network). So now, taxpayers are paying to keep Twitter/X alive. After all, it is taxpayer money going to the contracts the government gives SpaceX for launches. Nice way to subsidize what is effectively a one sided campaign machine for the GOP and far right.
> I suspect SpaceX will acquire Tesla at some point.
I think that is also likely, unless Tesla can stage a major turnaround, it is going to be beaten by Chinese competitors nearly everywhere that they are allowed (which is everywhere but the USA.)
This was my immediate thought as well. A great time to ask yourself — why am I literally paying for any of this? At best I literally don't use any of these services, at worst they are actively used against me.
I get what you're saying, but that taxpayer money is paying for the launch services at a very competitive rate (possibly the cheapest of all available options), not a subsidy scheme.
I guess the difference is Tesla is a public company, so requires more paperwork. SpaceX isn't public yet, but will be soon, meaning it will have a cash infusion.
This is fairly naive, Elon isn't the only investor in SpaceX.
My guess is "that they did the math" and had an engineering study which convinced them that getting AI datacenters into space will make sense.
It's also not hard to imagine why, the process alone once perfected could be reused for asteroid mining for example, then mirogravity manufacturing, either of which alone would be enormous capital intensive projects. Even if AI dataenters in space are break-even it would be a massive win for SpaceX and leave their competition far behind.
Are you a bot or are you just stupid?
There are several other companies that have announced efforts to try data centers in space.
I know this is hackernews and we like to get hyped up for new technologies, but, like, this just straight up isn't happening.
There is no benefit to putting data centers in space versus the giant cost that you would incur by doing so.
Can people please try and use their fucking brains for a second?
> Can people please try and use their fucking brains for a second?
Have you considered that people smarter than you think it is plausible?
> Have you considered that people smarter than you think it is plausible?
I know many people smarter than me, plenty of them who have spent careers building data centers, and not one of them think this is plausible.
You should consider whether people smarter than the average investor are pulling a fast one.
Maybe we are talking about different things here?
I don't doubt spacex can fail at this.
I also don't doubt we are fairly close to making this plausible.
> plenty of them who have spent careers building data centers
Famously, plenty of people who have spent careers building rockets would swear that reusable rockets would absolutely never work.
>I also don't doubt we are fairly close to making this plausible.
Maybe you should doubt that. There's literally no reason to think this is plausible besides some hype merchants' say-so.
> some hype merchants
Excluding Spacex:
Nvidia, Google, China, European Commission, Blue Origin
And this being HN, a YC funded company has put a single GPU rack in space and demonstrated training a reasonable sized model on it.
But yeah, it's all hype, sure.
On the off chance you're sincere and not just heavily over indexed into Elon stocks:
It's trivial to understand why this is all hype if you pay attention to physics, as another commenter suggested earlier.
https://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_law
Assume you're radiating away the heat for a single B200 (~1kW), and the max radiator temp is 100C, you find A = ~3m^2.
So that's 3 square meters per GPU. Now if you take into account that the largest planar structure deployed into space is ~3k m^2 (https://investors.lockheedmartin.com/news-releases/news-rele...), you're looking at 1000 GPUs.
That's a single aisle in a terrestrial data center.
Cost to deploy on earth vs satellite is left as an exercise to the reader.
You are missing one important thing here.
You do not radiate all the heat away from a GPU, a modern GPU can run pretty hot. Also look up how this is getting better for the next generation of GPUs.
Maybe repeat your calculation with updated assumptions?
But even if you were completely right, your argument is that we can't do this tomorrow, yes I agree. Typical technology development cycles are about 5-10 years.
> You do not radiate all the heat away from a GPU, a modern GPU can run pretty hot.
Fascinating. Tell me more.
Where does the heat energy that isn't radiated away go?
>You do not radiate all the heat away from a GPU, a modern GPU can run pretty hot.
LOL. If you don't radiate the heat the spacecraft just gets indefinitely hotter (until it glows and the heat is forcibly irradiated). It's space, there's no fluid to provide convection.
Have you considered that people smarter than you are scamming you?
Yep, definitely being scammed by not dismissing things outside my area of expertise out of hand.
I wish I had your confidence about everything!
I am yet to see any actual numbers showing how the economics of this would work or compare to the cost of building traditional data centers.
Please come back to reality.
> Yep, definitely being scammed by not dismissing things outside my area of expertise out of hand. I wish I had your confidence about everything!
Instead you put your confidence in Elon, who has zero expertise in this area?
> Instead you put your confidence in Elon
No, I put confidence my ability to do a web search, pretty rare skill nowadays ;)
You'll see that none of these are Elon/spacex, hopefully?
https://medium.com/@cognidownunder/google-just-announced-the...
https://www.ycombinator.com/companies/starcloud
https://www.informationweek.com/it-infrastructure/lunar-data...
https://ascend-horizon.eu/
https://www.axiomspace.com/orbital-data-center
I often wonder that same thing about many hn commenters!
From the Big Short (movie)
Jared Vennett (narration): "In the years that followed, hundreds of bankers and rating agency's executives went to jail. The SEC was completely overhauled, and Congress had no choice but to break up the big banks and regulate the mortgage and derivatives industries."
"Just kidding. Banks took the money the American people gave them, and they used it to pay themselves huge bonuses, and lobby the Congress to kill big reform. And then they blamed immigrants and poor people, and this time even teachers."
I feel like without adding some commentary with these quotes this comment lacks enough info to see how it relates to the linked article.
Because this move is entirely financial engineering to hide losses just like the roll up of X in to xAI.
None of this has anything to do with business or innovation. Do you not immediately see that? Most of my friends reaction to this news was that this is so obvious it's almost funny (or actually it is funny, since most were laughing as they read the headline).
I'm curious how you could not understand the relevance of the quote unless you were aggressively trying to not understanding it.
I understand it now, after reading the thread. There's a reason for that.
I have not been following the machinations of X very closely. I don't have the corporate structure of Elon's empire in my head, nor do I have the Meta or Alphabet/Google hierarchies in there. I couldn't have told you about the history of xAI beyond that it exists.
So that's plain ignorance of something you consider common knowledge, but I don't, rather than "aggressively trying to not understand it." And that phrase is particularly grating btw.
> I'm curious how you could not understand the relevance of the quote unless you were aggressively trying to not understanding it.
Monet probably wondered how other people couldn't see purple in a haystack.
What are they hiding that wasn't hidden already? Two private companies making a private transaction.. there is no mandatory reporting now nor after this move
SpaceX investors want to cash out, which is why they’re going public. Elon Musk wants to dump his X/xAI bags onto the public markets by merging it with SpaceX.
Essentially means that SpaceX investors are bailing out Elon Musk.
Don't forget that a lot of US mil stuff is launched by SpaceX so in a very real way they are the prime defense contractor in space for the country. If the public offering doesn't work, Unc Sam'll bail them out. Wonder if Trump will want a stake in the company this time.
Because X and xAI are both losing money. X needed cash to operate, so Elon rolled it into xAI to use xAI’s cash to help fund it. xAI is likely burning egregious amounts money, but will have trouble raising more capital. By rolling it into SpaceX he further covers up the financial issues because SpaceX is actually profitable. He can then raise more capital without having to worry (for a while) about how awful the burn is…
I, by and large, have a strong dislike of Musk to put it mildly. The one thing I will give him, and I think this is his real gift, is he’s absolutely brilliant when it comes to raising capital. He has proven to excel at raising capital, and deploying it well, for extremely capital intensive businesses. I do however wonder if the chickens are coming home to roost because both X and xAI are extremely unprofitable.
I think it’s almost inevitable we will see Space X and Tesla merge. The conditions of that merger will, I believe, say a lot about whether this move was brilliant or batshit.
Tesla acquiring solarcity was the same thing over. It did not make sense. Then and it does not make sense now. But the distortion field is so great no one notices.
SolarCity and Tesla made more surface level sense just being in the same general vicinity since they're both fundamentally green energy companies. That made it easy to spin questions about the financials with some CEO-speak about synergy.
However, the way Musk has become less subtle with this tells a story. He got away with these shady financial dealings multiple times so he's now becoming even more brazen and transparent with this behavior. We have gotten to the point in which the spin needed to justify his moves is the physics-defying viability of datacenters in space.
The distortion field will keep growing as long as he keeps getting away with it.
Doesn’t Tesla have a large and profitable storage business now? Probably could have just built that instead of buying SolarCity.
Why are space data centres physics defying?
Likely the intended meaning here is that the practicality of space data centers goes against the physical realities of operating in space. The single most prevalent issue with operating anything in space is heat dissipation in that the only method of doing so is via radiation of heat, which is very slow. Meanwhile, the latest Nvidia reference architectures convert such ungodly amounts of power into heat (and occasionally higher share prices) that they call for water cooling and extensive heat-exchange plant.
Even if one got the the economics of launching/connecting GPU racks into space into negligable territory and made great use of the abundent solar energy, the heat generated (and in space retained) by this equipment would prevent running it at 100% utilization as it does in terrestrial facilities.
In addition to each rack worth of equipment you'd need to achieve enough heat sink surface area to match the heat dissipation capabilities of water-cooled systems via radiation alone.
Not physics defying, just economically questionable.
The main benefits to being in space are making solar more reliable and no need to buy real estate or get permits.
Everything else is harder. Cooling is possible but heavy compared to solar, the lifetimes of the computer hardware will probably be lower in space, and will be unserviceable. The launch cost would have to be very low, and the mean time between failure high before I think it would make any economical sense.
It would take a heck of a lot of launches to get a terrestrial datacenter worth of compute, cooling and solar in orbit, and even if you ship redundant parts, it would be hard to get equivalent lifetimes without the ability to have service technicians doing maintenance.
Their viability is what I called physics-defying. Without some drastic changes to our current level of technology, the added costs of putting something in space along with the complexities of powering, cooling, and maintaining it once it's there is just too much to overcome the alternative of just building it on Earth.
Cooling.
Radiative cooling is the only option, and it basically sucks vs any option you could use on earth.
Second, ai chips have a fixed economic life beyond which you want to replace them with better chips because the cost of running them starts to outpaxe the profit they can generate. This is probably like 2-3 years but the math of doing this in space may be very different. But you can't upgrade space based data centers nearly as easily as a terrestrial data center.
More details from a guy who has thought this through https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
How do you cool them? Getting rid of heat is one of the number one challenges on the ISS.
Without evaporation and convection, getting rid of heat is a bitch in space.
Which is why he's the GOPs bank now
Parent poster may have been thinking of other readers. I see it as you do, but it's a fair question.
But that is also just an assumption isn't it? Could this not also be related to the fact that they plan to launch a ton of servers into the sky to run in space and power AI? It would mean that their AI product would become heavily based on the services provided by SpaceX via launching all this.
But regardless, I think quotes like these should have some commentary around them as it helps create a discussion around whatever point they might be trying to make rather than having to make assumptions.
> Could this not also be related to the fact that they plan to launch a ton of servers into the sky to run in space and power AI
FWIW, SpaceX launched a Tesla roadster into space without first having to merge with Tesla.
That's a very disingenuous argument and you know it. Starlink is under SpaceX. Do you also think that is wrong then too? They are effectively doing the same kind of thing.
> Starlink is under SpaceX
Kuiper is not under Blue Origin, and there are no whispers of Amazon and BO merging. You're the one being disingenuous in suggesting that companies have to be merged to buy services from - or cooperate with - each other.
That's not what I said. Please stop using these kinds of disingenuous attacks.
Data centers in space have never been a thing and never will be.
Low orbit satellites providing internet across the world also weren't a thing... until they were.
The biggest problem with satellite internet was the costs involved, which SpaceX has pretty much solved.
Datacenters in space, on the other hand, are a terrible idea because of the laws of physics, which will not get "solved" anytime soon. But don't take it from me, listen to this guy with a PhD in space electronics who worked at NASA and Google:
https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
Check the authors history. They are both anti AI and anti Elon. I think I feel a lot more confident staying optimistic and assuming that the SpaceX and xAI team have done their research about this. I know a lot of people are heavily biased in this matter due to politically not liking Elon or not liking AI, but I also think it's fair to say these companies have many very smart individuals working for them. If they have come to the conclusion that this is viable, then I have much more faith in what they are saying over one guys opinion who is biased against them and saying it's a bad idea.
You're also passing these judgements without knowing their full plan. Maybe we only know one part of the plan and maybe other details have not been announced. They may have a much bigger plan for this than just the specific information we have.
Having previously criticized someone doesn't make your technical analysis biased. It just means you noticed similar problems previously. Conversely, "I used to support him so I'm not biased" is given unearned credibility when really it just means you were late to noticing the obvious.
Technical analysis most definitely can be biased due to political leanings. This is why there is the whole idea that research can often be bought and paid for to get the results you desire. Because they are biased with money. Certain ideas or theories of how things could be done could very easily be overlooked or excluded by someone trying to dig for reasons to say something won't work.
What I am saying is that clearly SpaceX/xAI feel that this is a viable option based on many experts research/facts that are more knowledgeable than a single bloggers opinion. If I am thinking rationally why would I choose to believe a single random person over a group of experts banking A LOT of money that they have a solution that works?
It doesn't matter what their perceived, by you, biases are. WHAT ARE YOU GOING TO DO WITH ALL THAT HEAT?
I guess you'll have to wait and see what ideas they have to deal with this. If they can't manage the heat they aren't going to spend billions launching these things just for fun.
Why would they launch data servers into space? What purpose would that serve?
It’s hopium for his investors. Just like his robots.
Is it financial engineering or social engineering?
He's all over the Epstein files and his daughter has publicly verified that the timing works out and the emails are probably legitimate.
https://www.threads.com/@vivllainous/post/DUMBh2Vkk8D/im-jus...
At these scales, financial and social are very intertwined, it's both.
What are you talking about? They are both private companies. They don't have public financial reporting.
didn't tesla just 'invest' 2bills in xai?
yeah, I am not a huge fan of Musk, but this move is just going to bring down arguably the only decent thing he's produced.
Leave SpaceX alone you child. Gwynne has it in excellent hands.. find some other way to pay for your juvenile brainfarts.
Although I'm sure SpaceX would be a non-trivial loss, the most important idea - their truly reusable rocket -- is proven to the point where other people are assuming they should do that to make rockets, it's like if Benz' company goes bankrupt in 1899. In that universe the Mercedes probably never happens but the automobile idea is already a done deal.
What do you mean? SpaceX didn't invent the reusable rocket, and my understanding is that Falcon 9 is still not significantly more economical than disposable rockets, and that the main reason it's attractive is that it's not Soyuz-2.
Your understanding is wrong; see page 2 of https://ntrs.nasa.gov/api/citations/20200001093/downloads/20.... That’s a log plot!
The backing table is on page 8. Falcon 9 is (was, in 2018! It’s only cheaper now.) at $2700/kg to LEO. No one else is below $4k, except… Falcon Heavy.
> SpaceX didn't invent the reusable rocket
There isn’t a single inventor and reusable rockets emerged through decades of research.
But: SpaceX was the first to make orbital-class reuse routine and economically viable.
I found that surprising, so I looked on Wikipedia.
Soyuz-2 capacity to LEO: 8,600KG
Falcon 9 capacity to LEO: 22,800KG when expended, 17,500KG when not.
Soyuz-2 Cost to Launch: $35 Million
New Falcon 9 Cost to Launch: $70 Million
Used Falcon 9 Cost to Launch: $50 Million (cost to SpaceX: ~$25 Million)
Soyuz-2 cost per KG: $4000 (data from 2018)
New Falcon 9 cost per KG: $964 when expended, $1250 when not.
Use Falcon 9 coster per KG to Customer: $893 when expended, $690 when not
So realistically, Falcon 9 is roughly 20-30% the price per KG when new, and dropping to a minimum of 17.25% of the price when used.
Plus you get a larger diameter payload fairing and the ability to launch a payload up to 4X the size.
I'm pretty sure that even used as an expendable rocket, 1/4 the price per KG (if you need the capacity) is a pretty significant improvement. Now I understand why satellite ride-shares are so popular!
Plus, the Falcon launch cadence is infinitely better than Soyuz 2. 2025:
Soyuz-2: 12 launches
Falcon 9: 165!
Space is basically half the cost of it's competitors on a per kg basis. And while previous experiments like the DC-X existed, SpaceX absolutely gets credit for the first operational reusable rocket stage.
And I say that as someone that despises Elon and the way he casts his companies as due to his personal technical genius.
>And while previous experiments like the DC-X existed, SpaceX absolutely gets credit for the first operational reusable rocket stage.
Not true. What about STS?
Eh that's a spaceplane and solid rocket booster shells which I see as categorically different, and an absurd failure on a cost per kg basis.
It was a spaceplane and also a rocket. It literally had fixed rocket engines and was carried up by separating rocket stages. And yeah, it was expensive to operate, but it was built in the '80s and it truly was the first reusable rocket regularly flown, rather than being merely an experimental craft.
I think we're around stage 4 of:
I think he's genuinely changed for the worse, quite a lot, in the last 10 years. Staring down failure seemed to keep the worst tendencies in check, being untouchable amplified them.
>arguably the only decent thing he's produced.
such a hilarious comment / mindset. he made the best selling car in the world 3 years running. neuralink is a great breakthrough. there are a string of accomplishments which individually would be the greatest thing many many people have ever done.
> he made the best selling car in the world 3 years running
Not only did Elon not found Tesla[0], but many employees have described the "babysitters" or "handlers" who are responsible for making him feel like his ideas have been implemented, so that his caprice and bluster don't interfere with the actual operation of the company.
To give him his due, he's a phenomenal manipulator of public opinion and image, and he certainly has invested a lot of his emerald-generated wealth into numerous successful ventures - but he himself is not a positive contributor to their success.
[0] https://autoworldjournal.com/is-elon-musk-the-founder-of-tes...
I mean, even if he isn't directly making a lot of the decisions in these companies that are doing well, it doesn't mean he doesn't play a big role in that still. He still had to pick a lot of these leaders, pay them well, keep them satisfied enough to stay there, and also give them the proper freedom to lead these companies. There are many people out there who could also manage to make these companies fail instead of grow.
I feel that a lot of people simply don't like Elon because of political reasons which are often also based on misinformed opinions. It also can't be denied that he is an intelligent person. You can hear it when he talks in interviews.
Now I think ultimately any ultra wealthy person is going to have some flaws that people can find and latch onto in order to hate someone.
Trying to make a point out of whether he did or didn't found Tesla kind if defeats the rest of your post. He paid over 90% of the first funding round and brought in key people like JB Straubel. When the company was basically an incorporation paper and no assets. Under most companies people would have argued for founder/co-founder status at that point. So yes he didn't "found" Tesla but for all intents and purposes he basically did.
truly a braindead argument. probably paid for by a shortseller.
I was shocked to learn recently how China is crushing it in renewables and electric cars. BYD sold 600,000 more electric cars than Tesla in 2025, becoming the world's largest EV brand. Tesla's sales have been declining since 2023, while BYD sales are rapidly growing, so the gap is likely to get even larger in 2026. This is an important trend, regardless of how one feels about Musk.
Sources: https://www.science.org/doi/10.1126/science.aee8001 https://www.statista.com/chart/33709/tesla-byd-electric-vehi...
It's really disturbing how confidently blinded you are by whatever political biases you have.
So reveal the unbiased truth to us please -- what's the real motivation for consolidating these companies?
Hiding losses? From whom? He's the majority shareholder of both businesses. The combined company will go public and report on things like revenue, burn rate, etc. It's not financial engineering. It's a purchase.
Just say "rocket man bad" and save some keystrokes.
Maybe he likes the xAI minority investors more than the SpaceX investors? Or he needs their support for something else.
I agree we'll have to keep digging (or reading other comments, at least) to find a better explanation.
> Just say "rocket man bad" and save some keystrokes.
Hey Jeff, on what day is the wildest party on your island?
Legitimately, did you not immediately conclude it was for financial shenanigans? What did you think? I'm not trying to be shitty, but what else could there be?
Well if they plan to put a crap ton of new satellites in the air specifically for running xAI on it, I think there is a decent chance that it isn't purely financial shenanigans. Obviously the finances are probably a big part of such a decision, but companies also do these kinds of things all the time. I don't see why this is considered "shenanigans" or how the quote would relate to what is happening.
see https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
I'm gonna be honest, I really don't see how these opinion pieces are relevant. There are a lot of smart people working at these companies and I'm sure they have done a lot of research and work into determining that this is a viable thing to try. I am going to put more faith into that than somebodies opinion online.
It seems like a lot of people are very biased on this topic and want to see this fail because of who the company is. This author of this piece you linked appears to be both anti-AI and anti-Elon for example.
We also are unaware if there is some bigger strategy at play here and a bigger vision then what is currently being shared. I like to see companies try to innovate and take risks. I would like to try and be optimistic about things.
> I'm gonna be honest, I really don't see how these opinion pieces are relevant. There are a lot of smart people working at these companies and I'm sure they have done a lot of research and work into determining that this is a viable thing to try.
Like "robo"taxi, right? A lot of smart people have been working on this at same company for decade+
> I am going to put more faith into that than somebodies opinion online.
There are opinions and then there are things you can review that are factual and based on laws.
> Like "robo"taxi, right? A lot of smart people have been working on this at same company for decade+
I'm a bit confused what you're trying to imply here. They have launched RoboTaxi's and recently have been removing the human safety monitors in them. Are you trying to imply this didn't take a lot of work from a lot of intelligent people?
No, I am trying to not imply but say that it doesn't work which is why the company is now pivoting away to "humanoid robots" and is slowly starting to stop making cars.
You're talking about two different things. Robotaxi's are different from the self-driving style features of a personal Tesla vehicle. You specifically said RoboTaxi but now are referring to a pivot related to their Tesla vehicles.
I am just talking about (non-existent and will never materialize or exist) robotaxi. not about “full” “self driving” features of regular teslas (I own one)
They exist and are operational. I am confused by what you are talking about. How can you say "non-existent" and "never materialize" when they are quite literally in operation as a service right now.
maybe I am a fool, does space-based AI make no sense at all?
Literally none. Space is the worst possible place to put something that overheats already on earth. There's probably some synergy in the other direction (AI piloting of satellites or whatever) but that's marginal at best.
I've sure they've considered that in the engineering. For example, the solar panels would shade it. The space station has a cooling system in it. Musk's Starlink satellites don't seem to be overheating.
The problem is not shading them from the Sun. And the starlink satellites run at about 1 kW
good read: https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
that's an amazing read, lots of concrete and convincing challenges; but otoh, technology is evolving at such a fast pace, maybe it is possible for breakthroughs that we couldn't imagine now to become reality sooner than we would have anticipated?
It is a good read. Thank you.
There's another way to look at it, though. If the data center satellites can be built and launched cheap enough, you can still come out ahead on performance/cost. I.e. if the space data center has 1/10 the performance of a ground one, and they can be built and launched for less than 10% of the cost, then you've got a business. And there are costs that won't be incurred - no electric bill, no cost for land, no charge for maintenance.
I wouldn't be too quick to dismiss Musk.
* no electric bill: if you use solar panels to provide your own power, you also have no electric bill on Earth.
* no cost for land: land in sunny places where crops don't grow (for instance) is good for solar power and very cheap compared to building out a datacenter
* no charge for maintenance: sorry, I really don't get this one. Why don't the computers in space need any maintenance?
Why do your magic space computers not require maintenance?
So the whole space based data center thing is just a gimmick?
A gimmick, in a highly-financialized field? Surely not!
> maybe I am a fool, does space-based AI make no sense at all?
I think it does, for what it’s worth if we are to extend intelligence (as we know it) and potentially consciousness out there into the galaxy.
Because of distances and time, it is unlikely that humans will populate the galaxy with biological offspring (barring some technical breakthroughs that we have no line of sight on).
AI, on the other hand, could theoretically populate the galaxy and beyond, carrying the human intelligence and consciousness story into the future.
In, perhaps, a few hundred years.
Not sure if I feel comfortable with Mecha Hitler being our representative to the rest of the universe.
No. Imagine if your computer was in space instead of being under your desk. Would that solve anything?
Orbit is a very inconvenient environment. It's difficult to reach so maintenance is a nightmare, it's moving all the time, there's nowhere to sink waste heat into, you have a constrained power budget, you have a constrained weight budget. The only things you want to put in orbit are things that absolutely can't go anywhere else.
it absolutely doesn't unless there's a magical unobtainium cooling tech Musk got his hands on
Watts=heat
Heat has nowhere to go in space. Read about how much engineering went into cooling the ISS and now multiply that by billions.
A lot of people who are a little bit ignorant think it's really easy to cool things in space because space is notoriously very cold.
Physics, it turns out, is slightly more complicated than this and it turns out vacuum is an incredibly good insulator and more (much more) than offsets the temperature differential in terms of how easy it is to cool something.
Thinking a bit, ORBITAL ai makes little to no sense, nowhere to dump your heat, your gpus are going to be slag or only operate part of the time. But what if he put them on the moon? the lag time is what ~1.2s? That seems like an amount of time that a current AI query can take and still seem reasonable.
Not that I think it's anything but him allowing some investors to cash out when spacex goes public. Hell didn't he just shift 2billion from tesla to xai?
At the end of the day he will never see whatever bullshit he's peddling in the media about this sale his drug habit is going to kill him before then.
Does the moon offer much heat dissipation potential vs. orbit? The lunar surface seems like an almost-as-harsh environment.
Also 1.2 seconds is like ridiculously long, unacceptable latency.
You don't see how the common thread is Elon Musk buying out his own businesses, largely on the basis of overinflated stocks and corporate welfare?
It's baffling that the market has stayed so irrational because of Musk. It will collapse because of him.
SpaceX does real work at a profit, and its competitors will need even more time to catch up than they did Tesla.
Obviously there's a pattern of financial engineering, and it's inefficient, but the winners do offset the losers so there won't be a total collapse.
without having watched the Big Short or having read the article, my first impression from the quote is "Megacorporations are failing dramatically, and the billionaires at their helm are freely doing financial gymnastics to pull the covers over the eyes of shareholders, while gaming the system to fully circumvent taxes and regulation -- the people with the power to do anything about it (legislators and regulators) watch idly (maybe profiting), the oligarchs make off like bandits despite copious failures, and the end consumer/taxpayer is either robbed or clueless this is going on, but most likely both, when there was a world where accountability could have been had and the common man was treated better."
the article headline immediately screams "financial gymnastics" to me so the rest followed from the quote.
I see what you're saying, but I also know that companies do these kinds of things all the time. It's very normal for companies to move around like this if it results in better financials. I don't see how that really makes this "financial gymnastics". Pretty much every company out there does some funny things with the numbers in order to reduce their tax burden. I wouldn't doubt this is the same kind of thing. If xAI plans to launch a ton of servers into the sky, it kind of makes sense for them to be apart of a company they also own that just so happens to launch satellites.
Starlink is also a company under SpaceX. Would you argue that is also financial gymnastics? Is it much different from what Starlink does? Instead of launching satellites to be a world wide ISP, they are launching them to be an AI provider.
I just don't see how this compares to the quote, otherwise it would apply to so many companies, including other ones already under SpaceX.
To me this just doesn't seem related and seems like a pretty big stretch likely biased by people who dislike AI and Elon.
This merger smells like a bubble. Servers in space? They don't make enough to cover costs here on earth. Americans will be forced to bail this mess up because we need Falcon 9, Starship one, etc.
The military (and/or government) should keep paying in advance for anything they need from SpaceX and make sure other unsecured creditors are not tooo significant.
When it all goes bankrupt, they can pay off the bonds for x¢ in the dollar and own SpaceX.
Perhaps if the gov could organize a little better, they'd make sure SpaceX owed lots of taxes and put themselves in front of the queue for ownership and screw other creditors (especially foreign).
Edit: looks like the US military doesn't spend that much on SpaceX: https://londoneconomics.co.uk/blog/publication/crouching-riv...
The magic is not money or subsidies. Boeing got far more and they produced bupkis. Its organizational excellence. I don't know if that would survive. I guess if they can keep Falcon 9 stable then its still worth something but I imagine the star employees who grind themselves to dust getting this stuff to work do it for the mission and would depart if this occurred. Would Falcon 9 fall apart of that happens?
> Americans will be forced to bail this mess up because we need Falcon 9, Starship one, etc.
Or they could just go with the competition. If it came down to propping up something, I don’t see much difference between propping up ULA, Blue Origin or SpaceX. In the current environment who gets propped up probably depends on who scratches Donalds back.
This could just be a gamble they hope pays off. Or maybe there is some kind of other bigger picture strategy at play here that goes beyond just the AI from space idea. I try to stay optimistic about innovations in tech and I like to see companies willing to try new things instead of just staying stagnant.
For example, I think the car market had become pretty stagnant with traditional car makers, and most electric cars they attempted to make sucked. Tesla making good desirable electric cars really pushed EV's into becoming more popular and having a better charging network. I think it would have taken much longer for EV's to start growing in popularity if someone wasn't willing to take a risk.
Are they going to be too early to the market for this kind of tech? Maybe. Is it going to end up being a waste of money? Yeah it totally could be. But at the end of the day I do like to see some risks being taken like this and it sucks seeing constant negativity whenever companies try something new.
It is pretty clear. Socialize the losses of Musk investments by recovery via SpaceX contracts, supporting the US space program and the new Golden Dome program.
He sold FSD for 12 years, now is going to sell a Dyson Sphere for the next 30. This guy makes Ponzi look like a street hustler.
It's messed up that Grok underwrote all those subprime mortgages in 2008
I think the argument is that it's messed up that a large debt swap from xAI kept Musk's margin on Twitter from being called by his investors, and now that debt is being absorbed by SpaceX.
> Musk's margin on Twitter from being called by his investors,
Primary and largest investors in X are: Elon Musk, Saudi Prince Alwaleed bin Talal, Larry Ellison, Jack Dorsey.
I don't know that you need to worry about their financial well-being or that they are getting a raw deal.
I think people are more concerned about SpaceX getting the raw deal here.
And specifically that if the music is about to stop SpaceX has an implicit government backstop
It doesn't have to; the government's rescue of GM in 2008 killed a bunch of brands that they owned.
But given the current administration, I don't have a lot of faith in the government looking out for anyone else's interests here.
And TARP destroyed 4 of the 5 largest investment banks in the US, but it still left a bad taste in a lot of people's mouths
Starlink is about to get billions and billions from the BEAD program, on top of this.
> SpaceX getting the raw deal here.
Have they complained?
You’re really asking whether anyone at a private company is publicly speaking up against the famously emotional and vindictive owner?
Yes. People are saying they’re worried that the poor private investors of SpaceX are getting the short end of the stick.
That seems like misplaced concerned for an investor class that really aren’t suffering.
This thread specifically excluded the big investors, but they too have nothing but loss popping the bubble: Musk has been talking up the value of their investment. If they criticize in public, they’re just costing themselves money — much safer to sell and walk away.
Well, no, the worry is that xAI's bondholders, who are also Twitter's bondholders, will be indemnified from any loss on those bonds at public expense because they are now also SpaceX bondholders and SpaceX is a national security interest of the US.
I think unsavory business practices actually affect approximately everyone, even those not directly connected to any one particular instance of unsavory business practices.
Culture exists, after all.
Well this was just announced, and I'll be surprised if nobody gripes about a $2T dilution of their equity.
Yeah, the financial well-being of those investors is not what people are worried about here
Whoa, I had to do a double-take on the Dorsey mention -- like, didn't he take the money and run while laughing at the folks that overpaid? But it seems he's retained a 2.4% ownership stake in Twitter/X, according to Wikipedia:
https://en.wikipedia.org/wiki/Jack_Dorsey#Twitter
Still, don't make the mistake I did, which was to read the above comment to mean "he put more money in at the time of the buyout", since he was called an "investor in X".
Serious question .. are you long on cash or what? What investment class is not overvalued? Gold is but a store of value and doesn't really grow - look at how high it has gone.
I really don't know what you're trying to say. From your comment alone the conclusion I drew was that we should spend more on the industries that makes physical instead of financial products, such as SpaceX.
The sequence of events: Elon doing a leveraged buyout of X, then xAI funding, then debt transfers to X, then the xAI–X stock deal. Now the proposed SpaceX–xAI merger appears to have shifted X’s financial burden from Musk personally toward xAI investors and, potentially, future SpaceX shareholders.
This is speculative, of course, but yeah seems likely.
So what? People who buy SpaceX shares should take into account all of its debt when deciding how much to pay for a share.
Just like people should've thought about all the Related Party transactions when deciding how much to pay for Enron.
Are you suggesting there is straight up Enron style legally defined fraud in this deal? What is being hidden?
And yes, agency risk is always a thing. It’s part of life.
Won't someone think of the future SpaceX shareholders!
Won't everyone essentially be a shareholder indirectly? so yeah, someone should think about it.
SpaceX is a private company
Check back in 2 years to see if your statement is still true.
SpaceX is planning the largest IPO in history aiming for over a trillion dollars in market cap
All this is happening before then, no? So people can take that into account when pricing IPO shares, or deciding if the IPO ask is too high.
so are the banks?
>> SpaceX is a private company
> so are the banks?
Which relevant bank do you have in mind that is not a public company (listed on a stock exchange)?
Step 1: Merge xAI and SpaceX
Step 2: IPO SpaceX
Step 3: Merge Tesla and SpaceX x xAI (which would have been tricky if they were still private).
Where's the government bailout in this transaction that would make this a relevant comparison?
https://arstechnica.com/tech-policy/2026/01/starlink-demands...
Who is SpaceX’s biggest customer? And which industry are we being told by any number of governments around the world is the most importantly thing ever and must be subsidized and forced on people?
> Who is SpaceX’s biggest customer?
It is estimated that Starlink is, accounting for 70% - 80% of revenue. Sources: [1] and [2]
NASA is SpaceX's biggest external customer for rocket launch services.
Although NASA is SpaceX’s largest external customer for traditional launch services, the company earns far more revenue from Starlink customers (millions of subscribers). So overall Starlink itself is SpaceX’s biggest revenue generator and de facto largest customer segment.
[1] https://pestel-analysis.com/blogs/target-market/spacex
[2] https://londoneconomics.co.uk/blog/publication/crouching-riv...
From [2] (abridged):
SpaceX saves its biggest customer money by being the superior, cheaper launch option. The alternative was ULA, which was an extraordinarily expensive monster.
Please highlight the problems you have with how it pertains to this context, how the biggest customer is harmed.
What do you care if its private owners are willing to absorb the mess that is xAI?
It might be less about caring and more about pointing and laughing.
There has been rumblings of speculation that when the AI bubble pops that the government will bail the companies out.
Elon's usual modus operandi: take the thing that is losing money and merge it with the thing that is making money.
Prediction: at some point SpaceX will acquire all Tesla stock and take it private.
Why not put these AI/PV installations somewhere out in the ocean instead? A tiny fraction of the energy required to ship them there, you can actually physically get there to fix them, can use seawater for cooling, can use existing Starlink for connectivity, etc. Why/how is space more economical than international waters?
> In the long term, space-based AI is obviously the only way to scale.
I never questioned it.
Space is also extremely cold, and if it's as dense as Musk cooling won't be an issue.
I can't tell how many layers of sarcasm are here, but I just want to highlight that aktshually cooling in space is quite difficult because there is no convection, so the only cooling option is radiative. Which gets a bit hard when the satellite gets blasted by the sun.
The ISS doesn't have problems staying warm, it has problems cooling off.
> the only cooling option is radiative.
It does say he's planning an AI sun, I'm guessing that's the temperature you need to run at for radiation to work.
> It does say he's planning an AI sun
Everything I've heard from Musk in the past decade has been against my will and has made me dumber. (no I do not care to verify or know whether the above is true)
Edit: ah fuck ya got me "the next book in SpaceX and xAI's mission: scaling to make a sentient sun to understand the Universe" what the cultish bullshit is this. In a just world investors would be fleeing in droves from this cuckoo behavior (I know xAI & SpaceX are private)
There are already large communication satellites that consume several kW of power.
Oh, good. So we only need to multiply that by 200 million times, per space datacenter.
The data center would still consist of many individual satellites, much like a earth based data center consists of many individual servers
A large telecommunications satellite operates at about 15kW. A Blackwell GPU consumes 1kW so you would be at 15 Blackwells per satellite. The cooling surface needs to scale linearly so there is little return to scale.
This doesn't sound like a good idea to me.
tbh you could just combine them with starlink sats. didn't they just apply for (and get?) a license for 1 million sats? Stick a single racks worth of gpu power on those and hey presto you've just got yourself the largest ai cluster in the world by far.
Hence the “dense as Musk” comment
> scaling to make a sentient sun to understand the Universe and extend the light of consciousness to the stars!
I didn't realize SpaceX's media press is even cringer than Elon's average tweet...
There's an audience for this. People have been taking these vapid statements and his trouble with communicating as evidence of genius for well over a decade now.
Microdosing might not be the word for it.
I mean this is clearly directly written by Elon. I suspect the SpaceX comms people are equally eye-rolling.
My question is how they'll update the hardware, if AI processors get outdated every 4-5 years? Or they will keep outdated satellites just for routing and Internet-related services?
I still don't understand the "data center in space" narrative. How are they going to solve the cooling issue?
Maintenance cost must be pretty fucking insane
This "Space Datacenter" sounds like biggest bullshit in last decade, which is pretty damn fucking high bar.
Hes committed to building thousands of Optimus robots for a market that does not exist while cutting back on building evs (a market that does).
I think its pretty clear that Musk has lost his goddamn mind. And the American corporate system and Government seem powerless to do anything.
The reason is probably that Tesla is falling behind on EVs, or at least feels like they've juiced all they can from them at the moment, but advanced robotics is still on the upswing and probably is far from reaching its full potential. They have enough money that moonshots like these probably seem irrelevant at their scale.
As for the space datacenter idea, I think this is just a case of extreme marketing that Musk's ventures are so accustomed to. Making huge promises to pump their stocks while the US government looks the other way. When time comes for them to deliver on their promises, they've already invented ten more outrageous ideas to make you forget about what they promised earlier. Hyperloop as a viable mode of transportation, tunnel networks for Teslas, SpaceX vehicles as a mode of transport, X as the new 'everything app', insane timelines for a Musk-led human mission to Mars. They've done it all.
To be precise: humanoid robot TAM $0; vehicles TAM ~$2.7 trillion.
There is no maintenance, you have many cheap satellites - if one fails you just deorbit it.
How many is "many"?
One million.
You think I'm joking but I'm not. https://spacenews.com/spacex-files-plans-for-million-satelli...
I think it's fair to say past 30 years. Dotcom boom only had modest cons by contrast
Same way the dude solved Roadster, full self driving etc.
Presumably the cooling problem gets hand waved away as a technical detail, and the real selling point is data centers that aren't subject to any regional governments laws.
> and the real selling point is data centers that aren't subject to any regional governments laws.
No? ISS isn't exempt from legal systems.
https://www.esa.int/Science_Exploration/Human_and_Robotic_Ex...
Your link explains that the legal system arises from government agreements specific to the station.
So the csam generator can live on? Why not just pirate radio style it and cool it with ocean water....
Send up a spacecraft with back-to-back / equal area solar panels and radiators (have to reject heat backwards, can't reject it to your neighboring sphere elements!). Push your chip temp as much as possible (90C? 100C?). Find a favorable choice of vapor for a heat pump / Organic Rankine Cycle (possibly dual-loop) to boost the temp to 150C for the radiator. Cool the chip with vapor 20C below its running temp. 20-40% of the solar power goes to run the pumps, leaving 60-80% for the workload (a resistor with extra steps).
There are a lot of degrees of freedom to optimize something like this.
Spacecraft radiator system using a heat pump - https://patents.google.com/patent/US6883588B1/en
Periodically send a crew to a comet to bring back a large slab of ice to put in the data center.
beltalowda!
Remember MoviePass, and how they were losing gobs of money by letting people see unlimited movies for $20/month?
It was so obviously stupid that a bunch of people went, "well, this so clearly can't work that they must have a secret plan to make money, we'll invest on that promise", and then it turned out there was no secret plan, it was as stupid as it looked and it went bankrupt.
The "datacenters in space" thing is a similar play: it's so obviously dumb that a bunch of smart people have tricked themselves into thinking "wow, SpaceX must have actually figured a way it can work!"; SpaceX has not and it is in fact exactly as stupid as it looks.
But it won't end the same as MoviePass until Elon dies; he will keep moving things around, propping up failures with VC, IPO, federal/state (taxpayer) and profit making business money.
The pre-IPO timing of this narrative, combined with Musk’s history of Tesla’s stock pumping, leaves no room for doubt. But, that has worked for Tesla, I’m pretty confident it will work for SpaceX, which will IPO for $1T+.
that's the best part! They don't!
My guess would be just a regular radiator and cooling system like a liquid pump. The only obstacle should be the vacuum. That said I don't have any hopes Elon has any understanding of any of it.
Cooling a datacenter in space isn't really any harder than cooling a starlink in space, the ratio of solar panels to radiating area will have to be about the same. There is nothing uniquely heat-producing about GPUs, ultimately almost all energy collected by a satellite's solar panels ends up as heat in the satellite.
IMO the big problem is the lack of maintainability.
> Cooling a datacenter in space isn't really any harder than cooling a starlink in space
A watt is a watt and cooling isn't any different just because some heat came from a GPU. But a GPU cluster will consume order of magnitudes more electricity, and will require a proportionally larger surface area to radiate heat compared to a starlink satellite.
Best estimate I can find is that a single starlink satellite uses ~5KW of power and has a radiator of a few square meters.
Power usage for 1000 B200's would be in the ballpark of 1000kW. That's around 1000 square meters of radiators.
Then the heat needs to be dispersed evenly across the radiators, which means a lot of heat pipes.
Cooling GPU's in space will be anything but easy and almost certainly won't be cost competitive with ground-based data centers.
Sure, but cooling a starlink in space is a lot more difficult than cooling a starlink on earth would be. And unlike starlink which absolutely must be in space in order to function, data centers work just fine on the ground.
This. There's no scenario where it's cheaper to put them in space.
I think there's a lot of a room for an energy play that will ultimately obviate the enormously costly terrestrial energy supply chain.
You can just use the cheap solar panels that were gonna be launched into space (expensive) and not launch them into space (not expensive) and plug them into some batteries (still, cheaper than a rocket launch)
You forget the "in 2 years" part.
It will cost less to put it in low earth orbit than it will be to purchase land for it at any reasonable location.
I think that it's not just about the ratio. To me the difference is that Starlink sattelites are fixed-scope, miniature satellites that perform a limited range of tasks. When you talk about GPUs, though, your goal is maximizing the amount of compute you send up. Which means you need to push as many of these GPUs up there as possible, to the extent where you'd need huge megastructures with solar panels and radiators that would probably start pushing the limits of what in-space construction can do. Sure, the ratio would be the same, but what about the scale?
And you also need it to make sense not just from a maintenance standpoint, but from a financial one. In what world would launching what's equivalent to huge facilities that work perfectly fine on the ground make sense? What's the point? If we had a space elevator and nearly free space deployment, then yeah maybe, but how does this plan square with our current reality?
Oh, and don't forget about getting some good shielding for all those precise, cutting-edge processors.
Assuming you can stay out of the way of other satellites I'd guess you think about density in a different way to building on Earth. From a brief look at the ISS thermal system it would seem the biggest challenge would be getting enough coolant and pumping equipment in orbit for a significant wattage of compute.
Why would you need to fit the GPUs all in one structure?
You can have a swarm of small, disposable satellites with laser links between them.
Because the latencies required for modern AI training are extremely restrictive. A light-nanosecond is famously a foot, and the critical distances have to be kept in that range.
And a single cluster today would already require more solar & cooling capacity than all starlink satellites combined.
Because that brings in the whole distributed computing mess. No matter how instantaneous the actual link is, you still have to deal with the problems of which satellites can see one another, how many simultaneous links can exist per satellite, the max throughput, the need for better error correction and all sorts of other things that will drastically slow the system down in the best case. Unlike something like Starlink, with GPUs you have to be ready that everyone may need to talk to everyone else at the same time while maintaining insane throughput. If you want to send GPUs up one by one, get ready to also equip each satellite with a fixed mass of everything required to transmit and receive so much data, redundant structural/power/compute mass, individual shielding and much more. All the wasted mass you have to launch with individual satellites makes the already nonsensical pricing even worse. It just makes no sense when you can build a warehouse on the ground, fill it with shoulder-to-shoulder servers that communicate in a simple, sane and well-known way and can be repaired on the spot. What's the point?
Isn't this already a major problem for AI clusters?
I vaguely recall an article a while ago about the impact of GPU reliability: a big problem with training is that the entire cluster basically operates in lock-step, with each node needing the data its neighbors calculated during the previous step to proceed. The unfortunate side-effect is that any failure stops the entire hundred-thousand-node cluster from proceeding - as the cluster grows even the tiniest failure rate is going to absolutely ruin your uptime. I think they managed to somehow solve this, but I have absolutely no idea how they managed to do it.
Starlink already solved those problems, they do 200 GBit/s via laser between satellites.
And for data centers, the satellite wouldn't be as far apart as starlight satellites, they would be quite close instead.
No they didn't. 200Gb/s is 25GB/s, so... They could run 1/36th of a single current-gen SXM5 socket. Not even any of the futuristic next-gen stuff. 25GB/s is less than the bandwidth of one X16 PCIe3 socket. And that's already assuming the best-case scenario, and in reality trying to sync up GPUs like that would likely have loads of other issues. But even just the sheer amount of inter-GPU bandwidth you need is quite extreme. And this isn't some point-to-point routing like Starlink trying to get data from A to B, this is maintaining a network of interconnected systems that need to communicate chaotically and with uneven demand.
According to Gemini, Earth datacenters cost $7m per MW at the low end (without compute) and solar panel power plants cost $0.5-1.5m per MW, giving $7.5-8.5m per MW overall.
Starlink V2 mini satellites are around 10kW and costs $1-1.5m to launch, for a cost of $100-150m per MW.
So if Gemini is right it seems a datacenter made of Starlinks costs 10-20x more and has a limited lifetime, i.e. it seems unprofitable right now.
In general it seems unlikely to be profitable until there is no more space for solar panels on Earth.
All kinds of industries have been conserving more each decade since the energy crisis of the 1970's.
With recent developments, projected use is now skyrocketing like never seen since.
Before that I thought it was calculated that if alternative energy could be sufficiently ramped up, there would be electricity too cheap to meter.
I would like to see that first.
Whoever has the attitude to successfully do "whatever it takes" to get it done would be the one I trust do it in space after that.
His bet then, is that the $1 million cost to get a Starlink V2 mini into orbit can be made cheaper by an order of magnitude or two.
But it is always going to be significantly more expensive than a terrestial data center. Best-case scenario it'll be identical to a regular data center, plus the whole "launching it into space" part. There's no getting around the fuel required to get out of the gravity well. And realistically you'll also be spending an additional fortune on things like station keeping, shielding, cooling, and communication.
Elon prepping SpaceX for a meme-stock IPO?
Is there any other valid reason?
Datacenters in space is just stupid, getting rid of heat is much much easier on earth than in space.
Too me this smells of projected cash desperation. Do people actually pay for Grok?
You get Grok with paid x.com ; so there is some sort of cash from that, I would guess.
Is Grok driving any x.com subscriptions?
There are regularly comments on HN from people who say they pay for Musk's various products, and I am always downvoted into oblivion for suggesting that that is ethically problematic.
There's obviously quite a lot of autocratist illiberals in tech.
as you can see from the Epstein files, people with minor interests have a lot of money
I am concerned, and haven’t seen anyone else point this out yet, that Musk will move Grok’s CSAM generation capabilities to space to be beyond the reach of terrestrial policing. Does this create some sort of legal loophole here so Musk can do this with impunity?
No. For one, things in space are under the jurisdiction of the country they were launched from. For another, it's people that do crimes, not satellites or LLMs, and the people involved in making CSAM are all on Earth.
AFAIK he can do whatever he wants in space, but CSAM is still illegal to view or even download in most (all?) countries of the world. So unless the degenerates also move out into space (which I'm sure they're eager to do), it wouldn't really ease the legal situation here on Earth.
Ground stations would be the major problem.
Maybe if Elon launched himself and the dev team into orbit and didn't use any ground stations and just Starlink terminals he could start getting into legal loopholes.
So my floating data center in international waters idea has potential investors?
Remember that he wants to make X a bank. An orbital tax haven.
But seriously, I think legally satellites are under the jurisdiction of the country they were launched from.
Well, offshore launches are already a thing.
Or he could just buy a small island in the Carribean. There's one in particular that is available.
That island is part of the US. Technically SpaceX technology cannot be exported to other countries but laws are fake so...
5 days after Tesla gave xAi 2 billion.
Funnel money from a public company to a private entity and then make it disappear. Poof! Magic.
isn't this just fraud in broad daylight? I don't get it. Why not at least try to hide it?
It's crazy but legally there's nothing fraudulent here. I'm sure the deal was approved by the boards of both companies.
Its only fraud if poor people do it. Welcome to American politics.
But the "fraud" here is being done mostly to VC investors with deep pockets and lawyers, at least until he tries to take this entity public. And I can't imagine them just taking this lying down, but then again maybe they realize that offloading this steaming pile on public market investors is the best way out. But even then... SpaceX seemed like it was quite viable on its own, the investors there are the real losers here.
It is all very puzzling to me.
no I get it, but I mean fraud is usually kept out of the public. this is fraud in broad daylight?
Why hide it if you know you won't be punished for it?
It's only fraud if rich people lose money
Poor people are using their public car company to buy their private space company?
How do you keep these kind of things under the radar anyway?
There is literally an emoji in the middle of the announcement post. Very on brand for Elon.
Anyone remember the quote by Russ Hanneman on SV [0] - "No Revenue, means you're potential pure play"
We know datacenters in space - sound plausible enough - yet not practical - hence they're potential pure play - also you can have massive solar in space - unlimited space -- etc -- all true -- but how economical / practical is it ?
yet we know on earth - to power the whole earth with solar - only a fraction of the land is needed. Hell it's even in the Tesla Master Plan v3 docs [1] - current limitation being storage & distribution
so all you - are now witnessing to the greatest scam ever pulled on earth.
[0]: https://www.youtube.com/watch?v=BzAdXyPYKQo [1]: https://www.tesla.com/ns_videos/Tesla-Master-Plan-Part-3.pdf
The entire thing is a play. Musk should be a science fiction writer. He has that uncanny ability to create a statement that compresses 100+ years of industrial evolution into a few sentences.
>Factories on the Moon can take advantage of lunar resources to manufacture satellites and deploy them further into space.
I love how he goes from "the raw material is there" to "we will build high-tech supply chain to process them", just like that, magically.
https://i.imgur.com/wLJ60Vj.jpeg [I think you should be more explicit here in step two]
Also, https://xkcd.com/1724/
Edit: Formatting
Related: NASA chief suggests SpaceX may be booted from moon mission [1]. Blue Origin could snatch SpaceX's Starship lander contract. This looks increasingly a good idea.
[1] https://www.cnn.com/2025/10/20/science/nasa-spacex-moon-land...
Sean Duffy is no longer acting administrator of NASA. This proposal was apparently part of a bid to get the support of a coalition of old-space companies and new-space non-SpaceX companies. As part of that strategy he apparently leaked Isaacman's Project Athena document and was backgrounding that he was a SpaceX plant.
But, Isaacman is administrator now, and whatever you think about Isaacman and his relationship to SpaceX, I don't think there's much merit in thinking one of Duffy's half thought out plans is likely to be carried out.
Sadly this seems correct. When Trump was re-elected Elon Musk pushed for Jared Isaacman to be appointed as NASA administrator. When the pick went another way, it led to some real friction between Musk and Trump. Now, with Isaacman finally at the helm of NASA, it looks like Musk’s influence over the agency has come full circle.
> This looks increasingly a good idea.
Why?
The reason for the entire moon mission is national prestige.
Is financial fraud consistent with our national prestige?
There are better companies.
>Is financial fraud consistent with our national prestige?
You're right, that may be all we have left to show for it if people can't come up with something better.
Whether it's Musk or anybody else who's a real example of outright fraud, in a top position where honesty and straightforward dealing mean more than anything.
why must usamericans insist that they be the best at everything? it seems psychopathic...
The original Moon mission was masterminded by a literal card-carrying ex-member of the Nazi party (Wernher von Braun) and the American public back then didn't seem to mind.
All rocketry was, back then. You wanted ballistic telemetry? If you didn't know someone who worked on the V-2, you had to launch your own sounding rockets.
I think the parent's point stands. There's a lot more pragmatic concern with the damage SpaceX could do in 2026, versus the damage Nazis could do in the 1960s.
SpaceX hits delays according to the article
And blue origin has no delays because they don't even launch anything.
https://www.reuters.com/science/blue-origin-launches-new-gle...
FT says for 250 billion:
https://archive.ph/NqhWj
Thoughts:
1. What in the circular funding? This feels more like a financing scheme founding it under X/Twitter and then spinning it over to SpaceX. I suspect some debt is disappearing or taxes aren't getting assessed because of this move.
2. The only thing harder than harnessing "a millionth of the sun's power" on Earth would be launching enough material into space to do the same thing. And that's not even a reason for SpaceX to own an AI company, at least not at this point. The current AI isn't going to help with the engineering to do that. Right now hiring 20-somethings fresh out of college is way cheaper and SpaceX has been very successful with that.
quick edit: dang, I even got point 1 backwards. xAI owns X/Twitter, and that means that SpaceX now owns X/Twitter as well as an AI company. Super suspicious that SpaceX could actually think that buying the social media part (a significant portion of xAI's value) would be worth it.
Grrreat! Grok in Space... now AI-generated non-consensual sexual materials can be made completely outside the jurisdiction of any earthly body!
Rah rah. Line goes up!
Seems like a way to put a lot of junk in space. If thats in earth orbit it will lead to a lot of junk falling from the sky in 10 years. If it all burns up that will be a lot of nasty shit in the atmosphere - millions of tons!
Datacenters in space are a terrible, horrible, no good idea: https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
Discussed earlier: https://news.ycombinator.com/item?id=46087616
The really skeptical take here is that eventually all of Musk's companies merge, or at least the biggest ones, for juicing that market value to get that $1T payout. Looking at Tesla.
Then he will spin off and then remerge again. He has several more small companies before he plays that game. Maybe next quarter!
https://www.spacex.com/updates#xai-joins-spacex additionally the longer article on SpaceX site
I thought xAi previously merged with twitter, so all of this is now rolled into SpaceX? Atleast the investors in xAi and the original financiers in twitter get a breather. SpaceX is the new bandaid for this hot mess. Let’s see if this ends up rotting SpaceX or if it gets healed.
In the past 40 years we’ve seen power per unit of compute decrease by over 40 million times. This says we need to put data centers in space because we can’t produce enough power on earth for AI. That won’t be the case as history has shown, but this is a great way to get AI money for your space ships if you’re going to IPO.
It's interesting as the space fans (who often dismiss AI) say this is a project to get SpaceX's space/Starlink profits for AI. But the AI people seem to think this is to get AI money for space projects.
Kind of a bad look - but I can't precisely say why. Maybe he thinks he can raise more capital this way than he could for each company separately? Especially raising more money for X might be quite hard - they seem to be quite a bit behind on the revenue side compared to OpenAI / Anthropic. With both companies merged he might just find enough retail investors willing to buy at sky high valuations.
Urgh... Elon is famous not for things he's done, but for saying things he is going to do. Do people still buy in to this? Elon Musk always promises things far in the future but doesn't make good on them. He hasn't succeeded in self driving cars. He is never going to mars. He is not solving the LA traffic problem with tunnels. His robots are the equivalent of the Metaverse. He's a phenomenal businessman, and understands that a story is part of that.
Accountants will be studying the deals and cyclical valuations of AI companies in the same way we study bank runs and FDIC insurance today.
Two extreme grifting companies owned by the same asshole join dark forces with each other.
One hand pays the other.
This makes me genuinely sad. SpaceX was the one thing of his that Elon has largely avoided screwing up. Imho, this is in large part due to Gwynne Shotwell. She seems to have the personality (not to mention, personal wealth) to kick Elon in the head when he tries to mess things up.
What’s happening now is nothing more than a transparent effort to couple the AI hype-wagon to SpaceX in order to drive the valuation higher in the minds of investors who still think that LLMs will completely transform society.
I’ll be thrilled if the rocket folks can avoid being distracted by this nonsense, but I’m not optimistic.
I’ve been following SpaceX since something like the 2nd Falcon 1 launch and this is the worst thing I’ve seen happen. Sad times.
I think it’s just financial, I don’t see this as being detrimental or disruptive to SpaceX much at all.
What a clever trick to throw more money (governmental subsidies) into a sinking ship (xAI and "AI" in general). Perplexingly this maneuver will probably boost stock prices thus creating more monopoly money to burn resources with.
[flagged]
Doesn't the idea of Orbital Datacenters imply that the constraining resource right now is physical space, and not compute, electricity, etc?
Did we suddenly solve the electricity problem, or the compute problem? As far as im aware there are still plenty of datacenters being planned and built right now.
Elon didn't want to get outshined after Sam Altman suggested that "maybe we build a big Dyson sphere around the solar system". When will people realize that these "geniuses" are only good at making money, and any benefit to society is coincidental.
This is very clearly a move to sell SpaceX IPO at overinflated prices by promising a vision of a future which is perpetually 5 years away. SpaceX is going to be the next meme stock.
They must have linked the wrong press release /s. I would have expected a press release about SpaceX acquiring xAI to talk about why they did that. Or at least mention xAI beyond the first paragraph. This is just Elon talking about space data centers
IF we develop beyond earth...then AI, robots and connectivity will likely be huge parts of it.
+ spacex already is the best way for many payloads to get to space.
+ starlink is already the best low orbit based connectivity solution.
+ x is already a great way to train virtual world AI.
+ tesla (and its robots) is a great way to train physical world AI.
+ space takes big $ and talent - this combo would have both.
the IF at the top is just that. but feels like an interesting convo for this crowd.
This is definitely better than merging with Tesla.
They can sell xAI/Grok to all automobile companies along with Tesla and other businesses(X.com included) just like the SpaceX services.
It would good to see how it was valued.
I'm not sure if this is meant to be sarcasm but is there really a need in the car market for on-demand CSAM? What actual use does Grok or any LLM have in a car?
With that number of more satellites in orbit, launching a manned rocket into space is likely to be too risky due the amount of more debris and satellites. And will it be net energy positive solution ?
One thing to keep in mind. xAI and SpaceX both have contracts with the DoD. So it makes sense he moved it there rather than Tesla. Not sure I buy the needing AI for doing more in space or if this is to save sinking ship, but if one of his two big companies needed to buy it to keep it afloat it makes sense it was SpaceX and not Tesla.
I'm wondering if SpaceX's going public will be delayed. If not we'll see the first test of the public's appetite for what the AI companies' balance sheets look like
Whenever computer chips go into space, they have to be hardened against radiation, because there is no atmosphere to protect them. Otherwise you get random bit flips.
This process takes a while, which is partly why all the computers in space seem out of date. Because they are.
No one is going to want to use chips that are a many years out of date or subject to random bit flips.
(Although now it got me thinking, do random bit flips matter when training a trillion parameter model?)
LLMs specifically are fine with random bits flipped for the results to be 'creative'.
That's not exactly how LLM temperature works. :). Also that's on inference, not training. Presumably these would be used for training, the latency would be too high for inference.
It doesn't work like that, but it can.
Latency would be fine for inference, this is low earth orbit, that is about 25ms optimistically. Well within what we expect from our current crop of non local LLMs.
not a problem for "AI". it's just a bit more spice (temperature) which Grok possibly need! jk!
Genuine question: is it even theoretically possible to find some way to dump the heat that would be generated by a "data center" in space?
Yes, satellites and the ISS successfully radiate heat today.
ISS solar panels only make 200kw at max.
https://taranis.ie/datacenters-in-space-are-a-terrible-horri...
It's not comparable to any data center.
> buy a dying social network for 44bil
> merge it with a company created out of thin air for 20bil.
> have a third company buy it.
put it back on the market for 1.5 trillion.
If investors are falling for it, I guess all we can hope for is that the government doesn't bail it out!
> My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space.
I have never been so tempted to join Kalshi
Speculation - Starship is a dead end all of this is a play to inflate IPO prices for SpaceX during the AI bubble. Then cash out and start a new private space company?
> Starship will deliver millions of tons to orbit and beyond per year
Excuse my naive physics, but is there a point at which if you take enough mass off of earth and launch it into space, it would have a measurable effect on earth's orbit? (Or if the mass is still tethered to earth via gravity, is there no net effect?)
Earth weighs about 5972200000000000000000000 kg. They are claiming to plan to launch ~5000000000 kg / year. That's 8x10^-14 %. You'd need some pretty accurate instruments to tell the difference.
So... Elon wants to literally build Skynet?
https://tinyurl.com/xai-joins-spacex
"Launching a constellation of a million satellites that operate as orbital data centers is a first step towards becoming a Kardashev II-level civilization"
So, basically give ourselves Kessler syndrome. Or is Elon trying to monopolize orbit entirely?
Let's call it for what it is, a payday for Elon. Paper billionaires have figured out they cannot cash out with out tanking their paper, so now you have these circular deals to extract as much as possible. If we had a functioning government they would step in and put an immediate stop to this on national security grounds.
Is there anything substantially different about Google's announcement https://news.ycombinator.com/item?id=45813267 that makes it any more sane than the Space-X announcement?
Didn’t Elon say that orbital solar collection was a stupid idea due to energy loss in transmission? Using AI as an almost proof-of-work shows that it may potentially be more complex problem than previously thought. If we threw Bitcoin miners up to those satellites you could literally beam money down.
Elon investors should try buying a lottery ticket, it also lets you dream of the future while not providing returns.
Musk is moving value out of public hands and into his own. He overpaid $44B for Twitter, then rebranded it as an AI asset by folding X into xAI. He pushed Tesla to invest $2B of shareholder money into xAI despite shareholders voting no. Five days later, SpaceX acquired xAI, effectively turning Tesla’s cash into equity in a private company Musk owns far more of. Musk controlled every step, there was no real arm’s-length process, and he almost certainly knew the outcome in advance. Musk and his private investors get control, inflated valuations, and IPO upside. Tesla shareholders supply the cash, take the risk, and lose leverage.
If Musk and SpaceX are serious about putting 1 million datacenter satellites into space, then they are not serious about Mars.
You cannot simultaneously build and launch 10’s of thousands of Starships to deliver 1 million tons of equipment and supplies bound to Mars while also committing to launching 10’s of thousands of Starships to orbit full of satellites.
They would need to quadruple their launch rate, and half of those launches would be Starships bound for Mars, the vast majority of which would never return.
How many Falcon9’s have ever been built? It is incredible to say you can build that many rockets and use up that much fuel on any reasonable time scale. You might as well say the Tesla Roadster version 3 will be a Single Stage to Orbit rocket car.
Why would you put data centres in space?
Seemingly to prop up the value of your companies
Maybe this is beyond parody…
https://x.com/elonmusk/status/1005577738332172289
Self-driving single stage to orbit rocket cars. It can drive you to work and then go on a Starlink launch to LEO and be back in time to pick you up.
Add your car to the SpaceX fleet and get paid to own a Tesla!
Right, just to meet the most minimal of the scenarios for datacenters, someone upthread has calculated "launches every 9 hours, 24/7" as a minimum.
This is either insanely ambitious genius or pure shithousery. I guess we'll find out which one it is in 10 years
Given some of Musk’s previous statements, I think I know which one it is already.
well, Musk has been overpromising and under delivering for a decade (or more?), so it seems pretty clear this too is shithousery, albeit possibly ambitious.
I've yet to attain full-stack mastery in my job, but Musk has already attained capital stack mastery.
Does this mean the foreign software engineers in xAI are now subject to ITAR?
> with no ongoing operational or maintenance needs
How is this possible?
I suppose one of the ADR’s read something like “…who cares about bitflips, man. Isn’t AI all about probability?”
Knowing the insane level of hardening that goes into putting microcontrollers into space, how to the expect to use some 3nm process chip to stand a chance?
A trend at the moment is to just hope for the best in cubesats and other small satellites in LEO. If you’re below the radiation belt it’s apparently tenable. I worked somewhere designing satellite hardware for LEO and we simply opted to use consumer ARM hardware with a special OS with core level redundancy / consensus to manage bit flips. Obviously some problems will present for AI there… but there are arguably bigger problems with AI data centres like the fact that they offer almost no benefit with respects to the costs of putting and maintaining stuff in space!
Any details regarding valuations etc?
I have so many conflicting thoughts that I cannot properly articulate yet. I can say though, this is not going to end well for most, it is clumsily premeditated and starting to feel like dude is just trying to be a Neal Stephenson character.
Trails of those low orbit satellites wasn't bad enough.
Can't wait to see pictures of night sky ruined by... A data-center in the frame.
Just checking (genuine question) there wouldn't be a sneaky way to weaponize a million satellites in orbit around the Earth, would there? I can't imagine it wouldn't have ever been looked into.
I thought they were part of twitter
I don't see the demand for space being there, OSS is driving costs down and there are still plenty of hardware and algorithmic optimizations we haven't deployed yet.
I will not be left holding this bag. This is such financial engineering nonsense, and if we had any sort of regulatory controls this would never be allowed to happen - especially BECAUSE of national security reasons.
Whenever Elon Musk says free speech, I get the chills.
Do this math include the cost and weight of the radiators? Because it obviously can't work without big radiators, and I don't see them mentioned in the math?
"Thou shalt not make a machine in the likeness of a human mind."
And so it began. The seed was sent into space. All going according to plan.
This makes a lot of sense. The commercial launch business is not large enough to support all possible Falcon launches, so Starlink was created to take advantage of the low launch cost and vertical integration and is now a major profit center for SpaceX.
Starship launches are only going to make sense every 779.94 days (the approx 2 year Mars-Earth proximity). The rest of the time, the launches could similarly be used to deploy orbiting data centers for XAi/Grok etc. Brilliant move.
Banksters struggled to sell off Twitter notes. Did they get out intact finally?
What's to stop president AOC from pulling the clearances of everyone working for SpaceX?
SpaceX launches about 60% of National Security Space Launch payloads. The only other active launcher cleared for these missions is ULA's Vulcan Centaur, which has launched 3 times total, ~once every 8 months.
And? Donald Trump's presidency has made it clear that "this is bad for our country" isn't a sufficient argument.
I'm not really interested in getting in a political mudslinging contest; you asked what's to stop a future president from doing so and I gave a practical answer.
> Current advances in AI are dependent on large terrestrial data centers, which require immense amounts of power and cooling.
You know what's even harder to cool?
> Orbital Data Centers
I really would like to see a cost and cooling breakdown. I just can't see how you can do radiative cooling on the scales required, not to mention hardening.
I thought this was a troll by Elon, now I'm leaning towards not. I don't see how whatever you build being dramatically faster and cheaper to do on land, even 100% grid independent with solar and battery. Even if the launch cost was just fuel, everything else that goes into putting data centers in space dwarfs the cost of 4x solar plus battery.
Nobody knows cooling satellites better than SpaceX
I cannot really tell satire apart from genuine opinions anymore.
(But I do hope it was satire, if not, cooling satelites was/is a big issue and they only have very modest heat creation. A data center would be in a quite different ballpark)
Maybe so, but the actual SpaceX engineers are powerless to stop Elon running his mouth.
This is basic physics lol
Perhaps SpaceX incentive is to lie?
I hope all the Tesla shareholders understand that they’re about to get hosed.
Musks making Tesla seem like a good fit into the portfolio.
Has SpaceX figured something out related to photonic chips that dramatically reduces waste heat generation of compute?
The first M&A announcement I've seen in my entire life that includes a laughing emoji; maybe that's what it is!
Why?
Because it's another tool to move money on books and make it seem that spaceX and or xAi look good to investors when needed. That would be my guess.
Pre-IPO price padding. xAI is going nowhere but at least for now it has some value. Move it under SpaceX, bump up SpaceX’s valuation and therefore it’s opening IPO price. Then kill xAI and write it off.
These are no longer tech companies, to they are financial, and power, instruments of the billionaire class
> These are no longer tech companies, to they are financial, and power, instruments of the billionaire class
SpaceX has made numerous breakthroughs in reusable launch vehicles, human spaceflight, satellite constellation, and rocket propulsion.
SpaceX is the world's dominant space launch provider with its launch cadence eclipsing all others, including private and national programs.
Huge drag we allow it to be controlled by a drugged out nut job.
All as a side effect...
This is terrible for Space-X. They're doing a great job. Musk has left running it to Gwynne Shotwell, who really is a rocket scientist. Now Space-X has a AI business unit they don't need, a new money drain, and more attention from Musk.
Should have merged xAI into Twitter. A failure there would not be a major setback.
xAI bought Twitter a bit under a year ago: https://www.bbc.com/news/articles/ceqjq11202ro
Wait, does this mean Space-X owns Twitter/X now?
I suspended my disbelief and gave it a chance but I couldn’t hold it anymore after the emoji.
I know that per HN's guidelines we're supposed to be "kind and curious", and "reply to the argument instead of calling names". But with some texts, engaging with individual arguments loses sight of the more important bigger picture. So while unkind, the most "thoughtful and substantive" thing I think can be said about this text is:
The man's a moron.
I swear Prof G mentioned this exact same thing happening today
It was also foreshadowed in the Benedict Evans newsletter.
They both have X in their names, just imagine the synergies!
Is he also talking about moving X's servers (since xAI owns X) into space?
SpaceX has jumped the shark.
Friendly reminder for anyone that forgot - xAI acquired Twitter, so now Space-X is the proud owner of a dying social media platform that they overpaid for.
Any claims that this is about putting compute in space is just a non-sense distraction. This was absolutely about bailing Elon out of his impulsive, drug-fueled Twitter purchase.
The only question now is: when they try to go public, will they be punished for wasting so much money or not? My guess is: not.
At least there will be AI and Agentic stuff in mars.
Makes about as much sense as Twitch buying Curse about.. a decade ago?
You can only form one bubble. If spacex IPOs and rockets, Tsla will go to $10.
We need more 'moonshot promisers' like Elon going public. Come on AGI people, come on immortality people
Purely financial shenanigans. Nothing to see here, please move along.
In other words, he's cooking the books again.
People confuse being able to think big with being allowed to think big. With this much money loaned to him, protectionism and too many PR stunts, I don't think he delivered big enough, and he won't be able to at some point. This is madness.
I think there is one more possible framing for this.
Recently xAI has been in the news for Groq's revenge-porn-like "undress them" feature, which seems pretty legally questionable.
Musk has also been in the news for his own Epstein-related activities.
If he can move Groq and X into space, well, there's not very many age-of-consent or revenge-porn laws in space as far as I know, so maybe he'll be able to do some sort of legal leverage where the space data-center can produce otherwise legally questionable AI responses with impunity.
Sorry but who buys this bullcrap? Sure you can take a tesla and drive it in your bedroom to use its AC, because you make teslas, but you could also use just a window attached AC?
So they use a valid and valuable company to hide a giant dumpster fire company. To add to that, their best argument is "AI in space", which has some real "solar roadways" energy to it. I honestly don't know how any SpaceX shareholder could approve this.
So Elon has more shares when SpaceX IPOs?
Musk’s schtick lately has always been to find a challenging problem, point everyone to it and say “ I will solve it by September “, have the stock shoot up and make money. His first dis that with self driving, then twitter, then xAi, and now robots and data center in space. These last two will last him a decade as these are both challenging problems to solve over night.
Perfect timing to offload that debt onto the bag— I mean shareholders.
We just had an X8.1 CME event, I just want to point out that at any moment we could have an x40 (we had a carrington event already in 1859) or higher event and all those sats at low earth orbit would be fried and start hitting each other, if SpaceX keeps launching more it becomes incredible probable that we might hit the Kessler Syndrome, and we would legit lose access to Space for a WHILE, including all of what satellites entail.
Are we ready for that as a modern society or are we going to start enacting regulation against it? I'm sorry but people wanting internet everywhere does not justify we going back to the dark ages for a decade or more.
US is dialing back even on global warming. There is no chance current government would risk space superiority for some kessler-shmessler that nobody has ever seen.
Pretty terrible for SpaceX. Of course they paid a crazy inflated price for xAI in an attempt to cash in on the IPO. This just devalues SpaceX and exposes the investors to all the AI bubble risk.
> SpaceX has acquired xAI to form the most ambitious, vertically-integrated innovation engine on (and off) Earth, with AI, rockets, space-based internet, direct-to-mobile device communications and the world’s foremost real-time information and free speech platform. This marks not just the next chapter, but the next book in SpaceX and xAI's mission: scaling to make a sentient sun to understand the Universe and extend the light of consciousness to the stars!
I think Elon's taken one too many puffs of hopium
You'd think he'd have a pretty huge tolerance at this point.
So he doesn't want to go to mars he wants to make a big space chatbot?
> I mean, space is called “space” for a reason. [Face with Tears of Joy]
Plus I suspect the sun is already sentient no need to reinvent the sun
The next step will be merging SpaceX and Tesla.
Tesla has probably the most valuable shareholders on Earth. Over years of empty promises and meme status, the stock has pretty much purged all the level heads. So it's mostly deluded Elon sycophants giving placing their tithe on the alter of his sci-fi fantasy smoke and mirrors game.
In reality he will be dumping the debt of twitter and xAI (and maybe spacex?) on Tesla shareholders, and buoying that with the added layer of hyper that spaceX brings.
Now he just needs to work in crypto satellites down to users via all the new phones supporting satellite link to SpaceX. I kinda expected that one first. Distributed payment network outside of government control/oversight seems like something he would be in to.
What kind of financial engineering is going on here? Is xAI about to go bankrupt or something?
Probably. They are building the largest data centers in the world with very little revenue.
I asked Gemini for a two word summary and it wrote "financial engineering"
Grok: Integrated Ambition
ah, good. another heap of "exaggerations" (lies) to investors so that the creditors don't take elon out back just yet. hooray.
Reminder that space only allows for radiative cooling (since there is no air to absorb heat) so data centers in space are going to have massive cooling panels.
The Howard Hughes of our time. Soon enough he will start pissing in milk jugs.
xAI owns Twitter... So now space company owns Twitter? Wtf
What about security?
It's a CONSTANT stream of new ideas with no payoff at this point.
Hyperloop > Neuralink > Self-Driving Cars > Robotaxi fleets > Personal Robots > Orbital Datacenters > [insert next vibe shift]
At what point do people start to see the ever-shifting goalposts for what they are?
Hyperloop was never a company project, neuralink was a separate company, tesla is rolling out driverless robotaxis and fsd is amazing, robots inevitably are going to do the majority of work - there’s no real doubt about it is there?
Datacenters in orbit seem insane so idk we’ll see
> robots inevitably are going to do the majority of work - there’s no real doubt about it is there?
There's a lot of doubt that the AI and compute to enable that would happen on commercially relevant timescales.
Consider: "do the majority of work" is a strict superset of "get into car and drive it". The power envelope available for an android is much smaller than a car, and the recently observed rate of improvements for compute hardware efficiency says this will take 16-18 years to bridge that gap; that plus algorithmic efficiency improvements still requires a decade between "car that can drive itself" and "android that can drive a car". (For any given standard of driving).
And that's a decade gap even if it only had to do drive a car and no other labour.
You can't get around this (for an economy-wide significant number of androids) by moving the compute to a box plugged into the mains, for the same reason everyone's current getting upset about the effect of data centres on their electricity bills.
And note that I'm talking about a gap between them, not a time from today. Tesla's car-driving AI still has safety drivers/the owner in the driving seat, it is not a level 4 system. For all that there are anecdotes about certain routes and locations where it works well, there's a lot of others where it fails.
That said: Remote control units without much AI are still economically useful, e.g. a factory in Texas is staffed entirely by robots operated over a Starlink connection by a much cheaper team in Nairobi.
I appreciate you engaging, but I'm not sure power how would be the limiting factor. Assuming an average of 1kW of compute needed per robot (for reference, Tesla's AI4 is ~200W, rumors say 800W for AI5, nvidia B200 is ~1kW), that's nothing compared to the amount of energy we use for locomotion (a car eats like 20kW at 60mph).
Don’t forget LEO satellite internet! Oh wait….
Reminder that SpaceX has received an estimated $38 billion in government funding over the years, and all of its returns are going to a small set of private investors.
Socialized losses, privatized profits. As is the American way.
In other news, Kessler Syndrome: https://www.youtube.com/watch?v=8ag6gSzsGbc
Man the real story in a few decades is going to be whether SpaceX was onto such a kille business that it survived being used asma fiscal dumping ground for the losses being incurred by mismanagement everywhere else.
What if enemy or some anti-ai activities or etc attack it? How to protect it? It's just a too easy target.
It's just a dumbest idea ever if Elon truly believes it. I'm pretty sure he doesn't.
Wait so now SpaceX is the company letting you undress people? Why mix gold and poop? SpaceX plus xAI.
Don't forget to opt out of SpaceX's product Starlink using your data to train AI: https://www.pcmag.com/news/starlink-wants-your-data-for-ai-m...
From a technical point of view this doesn't make any sense.
From a finance and accounting point of view this makes everything more cloudy. Which certain types of people really like.
xAI to cover X investors, SpaceX to cover xAI, us American public investors to cover SpaceX since it cannot go under for strategic reasons. Ultimate grift.
Musk earns a $1tn payout when Tesla hits $8.5tn dollars.
I expect the next step in this series of moves is to turn Tesla into a SPAC & have it acquire SpaceX, bringing its valuation nearer that 8.5t.
Major grift vibes, inventing half-baked reasoning to justify massive valuations. If money wasnt so deeply entwined with politics at this point, this is the sort of news that would launch fraud investigations.
Terrible news for SpaceX.
What this tells me - xAI is essentially a failure, though at what level I'm not sure.
The way I read it, X AI is not really profitable and Elon's creditors/coinvestors asked for something tangible for their money, a.k.a shares in SpaceX, his only business that still has some solid foundation. The rest is emois.
Discussion on previous speculation: https://news.ycombinator.com/item?id=46814701
Musk always merges his companies when one is suffering:
Twitter/X in xAI
SolarCity into Tesla
xAI into SpaceC
I am just waiting now for Tesla to be acquired by SpaceC as it has run into issues.
The whole universe was supposed to be turned into paperclips, now it is being turned into graphics cards to produce images of barely legal girls on X.
And Musk keeps grifting about Kardashev 2 civilizations while his rockets do not even reach the moon.
If SpaceX goes public, that will rescue his xAi shares. I wonder how he will rescue his Tesla shares.
What metric does reaching the moon fulfill?
Nothing, save for advertising that you can. And Musk obviously can't, or he would have by now.
What does Gwynne Shotwell think of this. She seemed level-headed, but is she also batshit insane now by osmosis?
As a fan of space this makes me sad. AI Datacenters are complete nonsense. And binding all of this stuff SpaceX is idiotic. SpaceX could do so many great things for space and now its all messed up in the AI race.
“SpaceX is doing great but I want some of that AI investment money.”
When Elon Musk talks about benefiting humanity, remember that the one time he had unregulated power (DOGE) he used it mostly to cut benefits for poor people, and to push ideological agendas. His only agenda is self-aggrandizement, and this announcement is cover for passing the hot potato of Twitter debt around.
Whenever Elon, Anthropic or whoever mentions humanity, they don’t mean every 8 billion of us on Earth. They mean themselves, their mates and whoever they deem worthy of the posterity they believe only they have the merit to design.
I thought this wasn't viable due to cooling requirements - how do you cool massive amounts of compute when the only option is to radiate it into space - nothing to convect it with?
Also, the incredible amount of grift here with the left hand paying the right is scarcely believable. Same story as Tesla buying Solarcity. Board of directors should be ashamed IMO.
Yes. It is very cold up there but there is also no matter, or very little matter. So head conduction and convection don't work, it's all radiation. When we are learning to solve heat transfer problems in engineering school we are generally taught to neglect radiation, because it's effect on cooling the system is typically second or third order when compared to the to "big C's"
It would take roughly 5000 square meter area to cool a typical small data center heat output (1 MW). Not great, not terrible.
Apparently, OpenAI plan to build 250 GW of computing capacity by 2033.
To put that in space, based on your numbers, that's 1,250 square kilometers of cooling - an area roughly equivalent in size to Los Angeles
That's a lot of weight to launch into orbit
That is a very tiny amount of compute though.
Yeah but these hyperscalers are building data centers that are 100 or even 1000 mW
Cooling and maintenance (part swaps, etc.) are one of many obvious reasons why this is bullshit.
Doesn't stop grifters, tough.
in actual datacenters you often don't even bother swapping parts and just let things die in place until you replace whole racks
Not my experience at a hyperscaler, at least a while back. It definitely made financial sense to swap a small part to get a ~50-100k$ server's capacity back online.
Can we at least agree this is an awesome thing to try?
Way more exciting than spending $70 billion on VR like Meta did when we all just wanted to play games.
> the world’s foremost real-time information and free speech platform
What a joke.
For anyone too busy to go through all comments.
The overall sentiment of the discussion is *overwhelmingly skeptical and critical*. While a small minority of users defend Elon Musk’s track record of defying critics (citing reusable rockets and EVs), the vast majority view the "AI datacenters in space" proposal as scientifically unsound and economically nonsensical. Most commenters interpret the merger as a form of "financial engineering" designed to bail out underperforming assets (Twitter/X and xAI) using SpaceX's valuation ahead of a potential IPO.
---
### *Category 1: Technical Feasibility (Thermodynamics & Environment)*
The most robust debate focused on the physics of operating high-density compute in space.
* *The Cooling Problem:
* Numerous engineers pointed out that space is a vacuum and therefore an excellent insulator. While solar panels generate power, getting rid of waste heat from GPUs requires radiative cooling, which is inefficient compared to terrestrial convection (air/water). Users estimated that the radiator surface area required would be massive and structurally prohibitive.
* *Radiation & Durability:* Commenters noted that cosmic rays cause bit-flips and degrade electronics. Terrestrial hardware (like standard GPUs) would not survive long without heavy, expensive shielding, or would require "space-grade" hardening that lags generations behind in performance.
* *Maintenance:* On Earth, failed components are swapped; in space, a failed GPU or drive effectively bricks the unit or turns it into space junk.
* *Latency:* While Low Earth Orbit (LEO) offers better latency than Geostationary orbit, users questioned the utility of high-latency inference compared to fiber-connected terrestrial centers, particularly for complex AI tasks.
### *Category 2: Financial Engineering & Corporate Governance*
A significant portion of the thread analyzed the merger as a financial maneuver rather than a technological necessity.
* *The "Bailout" Theory:* Users widely believe this deal is designed to offload the heavy debt and losses from the Twitter (X) acquisition and the high burn rate of xAI onto SpaceX, which is viewed as Musk's most solvent and valuable company.
* *SolarCity Parallel:* Many compared this to Tesla’s acquisition of SolarCity—a move previously criticized as a bailout of a failing Musk-owned company by a successful one.
* *IPO Preparation:* Speculation suggests this is a play to juice the valuation of a SpaceX IPO by attaching "AI hype" to it, allowing early investors in the struggling X/xAI entities to cash out or convert their equity into SpaceX stock.
* *Conflict of Interest:* Commenters questioned the governance of private companies where one individual (Musk) controls the board and directs mergers that may benefit him personally at the expense of specific shareholder groups (e.g., SpaceX employees or investors).
### *Category 3: Economic Viability*
Users attacked the business logic of launching datacenters into orbit.
* *Cost of Launch vs. Land:* Even with the cost reductions promised by Starship, users argued that land and grid connections on Earth are orders of magnitude cheaper than rocketry.
* *Solar Efficiency:* While solar is more efficient in space (no night/clouds in specific orbits), users argued it is still cheaper to simply build more solar panels on Earth and use batteries than to launch infrastructure into orbit.
* *The "Million Ton" Claim:* Users crunched the numbers on Musk’s claim of launching "a million tons" of satellites, noting it would require an unrealistic flight cadence (e.g., launching massive rockets every few hours continuously).
### *Category 4: Musk's Reputation & Rhetoric*
The thread discussed Elon Musk’s history of promises versus delivery.
* *Skepticism:* Users cited a long list of missed timelines and unfulfilled promises (Hyperloop, Full Self-Driving by 2017, Robotaxis, Mars landings) as reasons to doubt the "space datacenter" timeline of 2–3 years.
* *Mockery of Language:* There was specific ridicule regarding the press release language, particularly the phrase "scaling to make a sentient sun," which many found to be "drug-induced nonsense" or "cultish."
* *The Defense:* A minority of commenters argued that betting against Musk has historically been a bad idea, citing the success of Falcon 9 and Starlink as proof that he can solve "impossible" engineering problems.
### *Category 5: Regulatory, Legal, & Ethical Concerns*
* *Jurisdiction Shopping:* Some speculated that moving AI to space might be an attempt to bypass terrestrial regulations regarding copyright, safety, or content generation (specifically referencing Grok’s lack of guardrails regarding CSAM/deepfakes).
* *National Security:* Concerns were raised that SpaceX is a critical US defense contractor, and merging it with a "chaotic" social media company and an AI firm introduces unnecessary risk and leverage over the US government.
* *Orbital Debris (Kessler Syndrome):* Users worried that launching millions of tons of disposable datacenter satellites would clutter low earth orbit, increasing collision risks and potentially locking humanity out of space travel.
### *Category 6: The "Why" (Strategic Speculation)*
* *Energy Arbitrage:* A few users attempted to steelman the argument, suggesting that if Earth's energy grid becomes the primary bottleneck for AI, space offers the only unconstrained solar power source, despite the cooling difficulties.
* *Vertical Integration:* Some noted this creates a conglomerate similar to Samsung or aggressive Japanese keiretsu, where the goal is total vertical integration of energy, transport, communication, and intelligence.
hmmmm
Elon Musk is a genius, but he’s a financing genius. Look at the long history he has of false promises supporting financing deals between his companies and you’ll see this for what it is, a cash injection and a lie to justify it. He did the same thing with a fake solar roof demo when Tesla bought the almost bankrupt Solar City. He also shifted resources from Tesla and SpaceX to support X in the early days. Even founding xAI outside of Tesla, when so much of its valuation was built on its AI capabilities, was questionable.
Who got the money? Hahah
When does the market realize this is all just a shell game and the emperor really has no clothes?
We saw this on a much smalelr scale a decade ago when one of Elon's companies (Tesla) acquired a second one of Elon's companies (SolarCity) because it was broke and owed a ton of money to a third one of Elon's companies (SpaceX).
Elon was forced to go through with his impulsive Twitter acquisition by a Delaware court, an acquisition that was not only secured by a bunch of Tesla stock but also a bunch of Qatari and Saudi royal money. He then mismanaged Twitter so badly Fidelity wrote down its value by at least 80% [1].
So what did Elon do? Raised even more questionable foreign money into xAI, diverted GPUs intended for another of his companies (Tesla) into Twitter and then "merged" Twitter into xAI, effectively using other people's money to bail him out from an inevitable margin call on his Tesla stock.
Interestingly, Twitter was reportedly valued at $33 billion in this deal [2], significantly more than the less than $10 billion Fidelity valued Twitter at. Weird, huh? With a competent government, this would be securities fraud that would have you spend the rest of your life in jail. And even with all that, $11 billion was lost on the deal.
So here we are and it's time for the shell game to be played again. Now it's SpaceX's turn to bail out the xAI investors.
And what is the argument for all this? AI data centers in space. Words cannot describe how little sense this makes. Launch costs (even if the Starship launch costs get to their rosy projections), cooling in space, cosmic rays (and the resulting errors) and maintenance. Servers constantly need parts replaced. You can just deorbit the satellite instead but that seems like an expensive way of dealing with a bad SSD or RAM chip.
[1]: https://www.cnn.com/2024/10/02/business/elon-musk-twitter-x-...
[2]: https://www.cnbc.com/2025/03/28/elon-musk-says-xai-has-acqui...
Making a "sentient sun" is the most bald-faced asinine, drug-induced nonsense that should be the complete destruction of all credibility to anyone who said it or typed it or works for anything here.
Financial engineering. Twitter under Elon became a dumpster fire of porn and hate and big banks were holding 13B in bonds that wouldn’t be worth the paper they were printed on for the company alone so he just links it with his only company that actually is doing something worthwhile…
Not sure how X which “merged” wit X (formerly Twitter) and SpaceX really matter or synergize but here we are. It’s all about the money being protected. And this Ketamine using wierdo is gonna be the worlds first trillionaire. Yay all of us.
> By directly harnessing near-constant solar power with little operating or maintenance costs, these satellites will transform our ability to scale compute. It’s always sunny in space! Launching a constellation of a million satellites that operate as orbital data centers is a first step towards becoming a Kardashev II-level civilization, one that can harness the Sun’s full power, while supporting AI-driven applications for billions of people today and ensuring humanity’s multi-planetary future.
Apparently optimus robots don't work and he needs to start his final grift, space datacenters, while his datacenters on earth are powered by gas turbines.
Most likely he's just trying to bury his epstein involvement where was exposed lying by his own daughter.
fool me once (Solar city) shame on you, fool me twice...
Now those dumb articles about AI data centers in space from a couple months back make so much more sense.
knowing elon, he will make this actually work, thus fully vindicating both the financial engineering and his arrogance!
Disgusting.
Can someone convince me that this is not a) pure horseshit b) a plan for Elon to sneak enough mass into orbit to hold the Earth hostage? If you can bring millions of tons of anything into orbit around Earth you can destroy civilization, or just France.
> This marks not just the next chapter, but the next book in SpaceX and xAI's mission: scaling to make a sentient sun to understand the Universe and extend the light of consciousness to the stars!
One of the dumbest things I've ever read.
Musk: "What do we have that OpenAi doesn't have?"
Musk aide while high: "sPaCe"
Slightly less high Musk aide: "But what is the synergy, where's the moat and how could that be done in practice and most importantly is there any limiting factor on Earth before we have to bring AI into.."
Musk : "SPACE!!!!!!"
It is incredible to think that the extremes of the stock market are actually pretty similar, pink sheets/cryptos and these mega companies are actually the same. News fueled pumps and dumps to win the cycle of hype of the week
They were all inside his head or bots replying to him on Xitter
Ahahaha, who got the money?
Going to be marked at some delusional valuation and at IPO retail bag holders are going to get absolutely massacred.
Like Musk's other great public company failures?
They are failures as far as products
This is comical give the model y has regularly been the best selling car in the world and the only EV to ever achieve such.
Tesla is marked at purely delusional prices as well. Total hopium on Optimus taking off while their core business craters.
Elon put everything he had into spacex and tesla. Against all odds, both worked out. He's now betting tesla on optimus. People are willing to believe him given his extremely successful track record of beating the odds. Is it guaranteed? No. If it was guaranteed it wouldnt be an investment. You take on risk for the chance at a reward. Calling it hopium is ridiculous--if there's anyone on the planet that has a track record of pulling off the near impossible its Elon.
https://x.ai/news/xai-joins-spacex wrong link?
Well surely this acquisition is above board. Nothing funny going on here, just good old business as usual.
> Well surely this acquisition is above board.
What makes you think it isn’t?
There's an epic conflict of interest here with Musk owning most of both companies. And they're in entirely separate fields, there is no plausible synergy here to be gained.
How can you have a conflict of interest if they're entirely separate fields? They have different interests, so where's the conflict?
You don't need synergies to justify a merger. They're often used as justification as in paying well above market price. But it has nothing to do with actual justification. You can just have a holding company of businesses
The conflict of interests here is the conflict between musks interests and the other shareholders interests
Yeah but who can be hurt by this, these are both private companies? So whose interest is his "conflicting" with? I'm sure the shareholders will raise it with him and/or bring a lawsuit if they aren't happy (they probably are happy).
he is literally going to launch datacenters into space to train ai so they are a little related
edit: these replies aren't going to age well
Yeah, I'm not buying that. I don't see how that could be any cheaper than regular datacenters. It might just be technically feasible, but launching stuff into space will always be more expensive than not launching stuff into space. And all those pesky technical issues like cooling might be solvable, but I doubt they're that cheap to solve.
You're right, but in this sense:
literally (adverb)
informal : in effect : virtually
Used in an exaggerated way to emphasize a statement or description that is not literally true or possible.
Ex: I literally died of embarrassment.
He says he is going to launch data centers in space. We should all know better than to take him at his word on that by now.
No he is not. It makes no sense from a physics standpoint or an economic standpoint. And even if they were, it wouldn’t require whatever this acquisition is.
No, he's not. And if he does, he's as big of an idiot as his detractors say that he is.
Oh man, I sure hope he disclosed that
Musk has a history of having one of his more successful companies buying one of his less successful companies. xAI bought X, and Tesla bought SolarCity
Musk is notorious for shuffling assets across his companies to make some financials look better. For example, shuffling Twitter servers (and then all of Twitter) under xAI.
Apparently SpaceX is the ultimate vehicle for buying Elon's shit that no one else wants. They're also buying thousands of Cybertrucks.
https://futurism.com/advanced-transport/spacex-buying-unfath...
Need any solar shingles?
my partner got shingles a couple years ago, it was a very painful experience!
(to be crystal clear, I am making a joke equating the failed SolarCity/Tesla solar shingles to the (generally considered very painful) Herpes Zoster manifestation also called "shingles")
And hey, look at that, pure coincidence, Tesla just gave xAI $2B! Now it's back in Musk's private pockets! How handy!
lol
Musk's involvement for one.
It's just "Elon zomg lulz" trolling for updoots. This place pretends to know better.
Well, at least this time both companies are fully private.
For the next four months, at least.
What's funny? Do you think the investors are against this? The investor's aren't idiots. I imagine the typical investor in Elon Musk's companies would approve of this sort of thing. So what's the problem? Besides, its a private company with Musk as majority shareholder in both. That's the beauty of private companies, you can just do things.
I wish more companies were private and ambitious. I'm tired of companies like Apple making marginal spec bumps to their phones and milking the same products for decades
> What's funny? Do you think the investors are against this? The investor's aren't idiots.
Any proof of that?
More than 70% of voting shares supported the package, very close to the level of support in the original 2018 vote. This excludes Musks share.
And consider that this is retroactive, meaning it's backpay. They're literally voting to give the guy $50b for work performed. He has a lot of confidence from his investors. And if there were issues, there would be lawsuits. Ironically the only lawsuits that get brought up, like the one about the pay package, are basically trolls, from a guy that had 9 shares.
Besides the parent is the one making a claim that something not above board is going on so burden of proof is on him.
Finally, it's a private company where Musk is the majority shareholder. He's moving money from one pocket into another, and any moves will be reflected in his attempt to raise money with the IPO coming this year.
Why do people online pretend not to understand?
Nothing in your argument is proof that the investors aren't idiots.
I can't imagine the world view you would have to hold to think that people who manage to command tens of billions of dollars to invest are idiots, just tripped over the money and just go off vibes.
Apple just launched their own silicon chips just a few years ago. They're very ambitious but still calculated.
> > I'm tired of companies like Apple making marginal spec bumps to their phones and milking the same products for decades
At least what Apple does is real not make believe like everything Musk claims , disappear boring Apple or even boring Microsoft, Oracle, IBM etc.
And the world would come to a screeching halt, disappear all of Musk companies and people would barely notice.
You seem to be eager to be sold dreams , that's exactly what vaporware salesmen like Musk hope to find on their path
The investors want to cash out, Musk needs lots of money to plow into his latest toy that so far only excels in ridiculing him and sexual harassment/CSAM, so they make a deal to take in xAI and go public. Win win.
> The investor's aren't idiots
citation needed.
It's widely reported that Musk is a majority shareholder of xAI and the controlling shareholder of SpaceX (close to 80% of voting shares). Not surprising that he would be looking to consolidate ownership under one entity especially if he perceives significant synergies (i.e., data centres in space).
Data centers in space are a hilariously bad idea. Where would the heat go? This idea is like the opposite of liquid cooling.
Shocked to see SpaceX buy the datacenter in space meme. Where does the power come from? Where does the heat go? Why add (high) launch costs to your buildout capex? Why add radiation as another risk factor to your already-unreliable GPUs? Am I missing something fundamental here...?
Money! Also power source is just solar - not too difficult. I don't think radiation would be too much of an issue either since they're in low earth orbit. Heat is probably the biggest problem. Or manufacturing & launch costs. Pretty silly idea anyway.
Aside from Elon Musk, there are a few other people with a lot of capital aiming to do the same thing. That means, either they are all wrong (possible) or this problem has been solved somehow and the solution itself is not public.
Google and Amazon are doing the same thing. Maybe it is a moonshot (pun intended), but Musk is hardly alone in the push.
https://www.wsj.com/tech/bezos-and-musk-race-to-bring-data-c...
https://www.nytimes.com/2026/01/01/technology/space-data-cen...
Not to mention the huge issues of cosmic rays. Sure, if the lifespan of the satellite is expected to be low, then maybe tolerable. But even then, how would this be financially viable?
I didn't say it was a good idea, just that if Musk perceives it's a good idea then it makes sense why he would want to combine the two businesses.
I think it's far more likely that he wants to combine his businesses to roll his really expensive, debt-ridden companies into one entity with the company that actually reliably makes money.
Only a person who is high as a kite can think thats a good idea.
Some guy on hacker news argued they could just use radiators.
Radiators might be a reasonably effective way to reject heat if you can run your AI machines at 1000 K or thereabouts.
Indeed. But it's also a hilariously Musky idea! Some moderate technical competence paired with sociopathy and an ego orders of magnitude too big, and voila, you get Cybertrucks, Hyperloops, Neuralinks, Teslabots, datacenters in space, and all the other garbage the man spews.
I cannot wait for him to one day be hit in the face by reality.
I have never understood how Data centers in space ever make economic sense, the payload, latency and many other issues make it difficult at least for the immediate needs
Latency isn't an issue with Starlink - the data centers are in low earth orbit, not in GEO
given the max bandwidth of a starlink sat is in the 100Gb on a good day range why would you want to limit a DC to less bandwidth than a single cheap fibre?
Also in LEO you're going to have reentry become more of an issue (starlink burning up in the atmosphere isn't some free garbage removal it will have a measurable impact on the chemical make up (assuming it even burns up and doesn't just squash more farm buildings), Power supply more of an issue and still have huge problems with heat and radiation.
You mean unlike Hyperloops, Cybertrucks, Teslabots, Neuralinks, and all the other insane stuff that moron cooks up?
I'm sure these comments will be rational and non-biased by political emotions /s
wow this is one of the most incredibly fraudulent things that ever happened in American capitalism and I'm not ignoring Enron or the mortgage meltdown. I'm speechless the US has given up on any semblance of law and order in matter of financial markets and this stuff can happen without people going to jail.
This means that Grok, Elon's politically labotomized involuntary pornography generator, and X (formerly Twitter), Elon's Nazi-adjacent propaganda machine, are now completely intertwined with SpaceX, a too-big-to-fail government contractor that currently serves as America's only reliable option for manned orbital spaceflight.
Anyone who doesn't see how broken this situation is isn't paying attention. This is how people like Elon, who want to seize as much power from the government as they can, ensure that the means for seizing that power are untouchable.
Anyone who has ever used Grok or X lately knows that both of these products are heavily manipulated to align with the political, social, and economic views of Elon Musk, who is increasingly boosting "white power" language and full-throatedly backing America's most nationalistic and authoritarian president to date.
This is just another consolidation of power, and it's deeply worrisome. Any integrity one may have hoped remained at SpaceX just vanished when they aligned their mission with that of these deeply problematic digital services.
And this is not even scratching the surface of what looks like a deliberate attempt to create Kessler syndrome by launching millions of cheap short-term satellites into orbit, or the rationality of putting datacenters into orbit in the first place...
Seems more risky to me; it takes only one Mercurial temper swing for SpaceX to be nationalized, and now Twitter and xAI are bundled along with it.
Besides the obvious, why do engineers with real skills still work for this guy?
People pooh-poohing space datacenters will obviously think this is a bad move. But Elon clearly believes space datacenters will work. Given that, and the fact that SpaceX will IPO this year, this acquisition was inevitable.
SpaceX and xAI would not be able to freely collaborate on space datacenters after the IPO because it would be self-dealing. SpaceX likes to be vertically integrated, so they wouldn't want to just be a contractor for OpenAI's or Anthropic's infrastructure. Merging before the IPO is the only way that SpaceX could remain vertically integrated as they build space datacenters.
How can you clearly believe anything Musk says at this point. It's not 'clear' at all what he actually believes and what he just makes up.
Space datacenters just so grok can undress women? this is the dumbest company on the planet
Most everyone here is just angry Elon opened the door for freedom of speech to exist online and started the process of uncovering the left's extreme corruption. They feel betrayed by him so he's enemy #1. Literally could cure cancer and he'd still be branded as evil by the left. Ironic as he's arguably done more for leftist causes (renewables, ev's, etc) than anyone else on the planet.
These are private companies, with private investors. If Elon really screwed all investors over and over again he wouldnt be able to raise money repeatedly. But he has no problem raising money. Because the claims that he's doing shady things or screwing investors is entirely unfounded. Same goes with the claims that he's an idiot or not that smart. If thats really the case, show me what you've accomplished with your genius.
It takes intelligence to recognize intelligence. An idiot cant recognize a genuis.
Unfortunately you led with the claim that Musk cares about freedom of speech, a claim so laughable it renders anything else you say unbelievable.
Elon's purchase of Twitter (now X) opened the door for conservative view points to be heard online. Prior to that, all other major platforms were censoring conservative views (with wide spread company leadership claiming due to pressure from Biden administration).
I regularly use X. I get blasted with political posts from both sides constantly on it.