[urams]

Very possible Elon is doing this to make give Anthropic better chances against OAI while he attempts to reshape xAI.

Also possible he sees infra as the future of xAI if he really believes in the value of space compute.

Hard to see this any of this as anything other than a bearish sign for Grok though.

[tristanj]

That's very doubtful given recent news that Colossus is running at 11% capacity and has hundreds of thousands of idle GPUs. xAI acquired too many GPUs and currently doesn't have enough customers to use them. That's why they are making compute deals with Anthropic and Cursor.

xAI is bleeding money and this compute deal with Anthropic will pay for all of xAI's capex ($25 billion) in 2 years.

[petesergeant]

> will pay for all of xAI's capex ($25 billion) in 2 years

This is a confusing framing; you pay down capex with profit not revenue, and there is presumably a high opex cost.

[ayende]

Nevertheless, you are still talking about 15B / year - when the initial investment was < 30B. That investment also probably covers things like power, salaries, etc.

So you are still covering all of that.

[petesergeant]

> That investment also probably covers things like power, salaries, etc.

That does not seem probable to me, other than for the initial build-out.

[jasondigitized]

free cash flow pays down capex

[shenberg]

11% MFU does not mean 89% of GPUs are idle, it means that they're using the GPUs ineffectively.

[reliabilityguy]

11% capacity is not 11% MFU. The first is about actually using the hardware for something in the first place, the latter is about how efficiently you compute. Different things.

[jbs789]

The post you’re referring to doesn’t even include “89%” or “idle”. I’m not sure I understand the point you’re making then rebutting.

[tfgg]

The original post of 11% is referring to leaked numbers about _MFU_, which has erroneously been re-reported as fraction of GPUs being used at all. The parent post is trying to correct this misconception.

[jbs789]

I see. Thanks

[SecretDreams]

This might actually be worse from a power costs standpoint.

[andriy_koval]

> xAI is bleeding money and this compute deal with Anthropic will pay for all of xAI's capex ($25 billion) in 2 years.

minus maintenance and energy cost.

[utopiah]

> space compute

Heard anybody remotely competent about space talk the topic? It's pretty much a literal laugh every time.

[Forgeties79]

It is absolutely baffling to me how frequently I hear people talk about this around me. There is no way this is happening anytime in the near future.

[tuveson]

But if the US isn’t the first to put an AI on the moon, we’ll lose the AI space race!

[paddy_m]

Mr. President, we must not allow a mine-shaft gap!

[Forgeties79]

To this day still my favorite line in the movie. It really just punctuates the whole thing

[__MatrixMan__]

You don't have to actually have an orbiting datacenter for the idea to work. You just have to convince enough people. Once you've done that, you can claim that regional regulations don't apply to your data because the data is in orbit. Its not like somebody is gonna go up there and catch you in the lie.

Out-of-regulatory-reach is what they'll actually be selling. It can be on earth, it just has to be sufficiently hidden such that you can claim that it's in space.

[estearum]

I don't think that's the purpose and if it is, it's obviously trivially solvable. The US can and does simply assert jurisdiction all over the globe. There's no reason it can't or wouldn't extend that to orbit.

That conversation goes like this:

US: Stop doing that thing with your data center

X: It's in space, you can't tell us what to do

US: Yes we can

X: <Say OK> or <Go to jail>

[everfrustrated]

The US already claims jurisdiction on all launches from US and NZ (Rocket Lab). Every one of these had to get approval from the US to launch.

[wongarsu]

Some of the proposals are laughable, some of the example calculations are. The idea of running AI training sounds extremely challenging. But the idea of inference in space doesn't seem absurd

The power budget of a starlink v2 mini satellite is estimated at around 20kW based on the known solar panel size. That also matches what a satellite of that size would roughly dissipate without dedicated radiators, using just some heat pipes to spread the heat evenly over the satellite's surface. There is nothing fundamentally preventing you from taking the same satellite design, remove most of the comms payload and instead put 15kW of GPUs there. Or about 10 GB200 including the CPUs, networking, etc that you need along with the GPUs

Now, do the economics work out for $300k worth of compute for each satellite, in an environment where maintenance is impossible and degradation will be higher than on the ground? Probably not right now, but in a couple years they might

[beAbU]

The starlink satellites have hall effect thrusters, and presumably beefy laser and radio comms systems that will account for a lot of the energy budget. Also, they are sometimes in the earth's shadow, a naive calculation says they have 2x the solar budget they need, in order to charge up batteries for shade-time operations. There is no where near 20kw of compute in a single satellite, and thus no where near 20kw of heat to get rid of.

Furthermore, xAI's colossus supercomputer is specced at 250MW. And this seems to be a number that'll just increase over the coming years with new bigger DCs.

To match this level of performance they will have to launch what, ~15k satellites _per_ equivalent datacentre?

Regarding cooling: you can't just cover the outer surface with pipes. You cant't dissipate the heat, you need to _radiate_ it away. You need to point that surface to the deep dark cold of space. If you point it to the sun, you will heat your satellite. Think a massive "reverse solar panel" that works with infrared. You need surface area, and loads of it.

I'm not saying this is impossible. Obviously elon will prove us all wrong because he's stubborn like that. But there is no way this will ever be economically viable when competing with terrestrial based systems.

[wongarsu]

The assumption behind "spread the heat over the size of the satellite" is radiative cooling. Admittedly my numbers were a bit off, you need to make the satellite a bit bigger, or use some of your solar panels. A starlink v2 mini has 8m² of area per side, so 16m² total. To dissipate 20kW to space, at a surface temperature of 80°C and emissivity of 0.85, you need about 28m² of space. So you need to increase size a bit, or add 12m² of dedicated radiators. A bit more to deal with the real live complications (the sun exists, the earth is actually warmer than space and covers a significant portion of the sky if you are in leo, etc.).

The actual Starlink V2 Mini is has estimates for solar generation that range up to 35kW. We need a bit more, but not much more.

My numbers are a bit optimistic, but they are in the right order of magnitude. Power and cooling are very achievable in the area of putting one server on one satellite. It's the "putting one DC on one mega satellite" ideas that run into feasability issues, and for inference those aren't needed. The economics are the bigger issue, and launch costs are only moving in one direction

[beAbU]

I think you and I are coming to the same conclusion. It's totally viable to have 20kw of compute in a satellite in space. Hell, you can probably get away with 100kW or more with some sort of megasatellite.

I'm pretty sure we'll see this within the next 5 years.

But deploying a megawatt scale DC in space? And then doing that multiple times a year?? I can't see how this is the only solution to this problem. I can't see how that will be the de-facto way of deploying DCs. Not for the next 100 years or more.

[weirdmantis69]

The terrestrial based systems getting protested, blocked, and demonized everywhere? Those ones?

[beAbU]

Plop these satellites on a massive barge in the middle of the ocean with 2x the solar and battery capacity to (to account for day/night cycles) and you have exactly the same solution, only it's multiple orders of magnitude easier to "launch", service and you can actually recycle the deprecated hardware. It will also solve the cooling problem that's associated with space.

The fact that elon's colossus data centre needs to burn jet fuel on site for power tells you everything you need to know about the viability of a solar powered space based solution.

[lovich]

For everyone that’s blocked there seems to be another popping up even when the local populace voted against it.

Even then, space is still a harder place to put a satellite than any rural area on the planet.

[ashley95]

How is this not absurd? What is the benefit? Space is a harsh environment, with issues due to solar radiation etc, etc. And it's permanently 100ms away from any user.

[sulam]

Tl;dr — but it’s too heavy

You’re making some decent points here, but you’re either forgetting or ignoring the major thing that people usually neglect to mention when they want to make a case for this (crazy) idea — weight. Unless SpaceX is going to completely redesign hardware such that it is optimized entirely for its mass, it requires many (many) launches to even get a small set of racks into space. I don’t normally get up in arms about the CO2 emissions of data centers, I think there is offsetting value created by their use, but I would absolutely protest trying to put data centers in space and do my best to shut down the hundreds, if not thousands of launches it would take to achieve even a tiny fraction of an AI data center.

[jpadkins]

Have you calculated the CO2 output of Terran datacenters run on natural gas vs. space datacenters run on solar? The launch CO2 usage is one time, the datacenter energy is for the life of the equipment.

[sulam]

A Falcony Heavy probably generates 1 kiloton of CO2 per launch. Data centers on the planet are highly variable depending on their energy mix. It's true that a large a datacenter running on natural gas or coal power is significantly more in a year, but the sheer number of launches required to get that same data center into space is actually comparable, and there's no saying that this is the end of it. Oh and we should also have questions about how you safely de-orbit these things.

[vel0city]

If solar+batteries are so cheap and viable to power giant 100 megawatt datacenters in space, why not do the same on Earth instead of using natural gas generators?

[jpadkins]

Solar panels on surface get ~70% less energy than in space once you factor in atmosphere absorption, day/night, weather, etc. Solar + battery on the surface is not economically viable versus nat gas generators (excluding externalities)

[lovich]

What could possibly change that would make it a more economically viable option than putting it on earth. Nothing about space improves how data centers operate, and there are several new issues introduced by being in space.

And that’s before you get past the cost of actually putting the shit up there instead of just driving or shipping by rail.

It’s just more looney tunes delusion from a guy who is just talking out his ass about shit he thinks is cool, like the cyber truck was.

[sebmellen]

While I agree with you on the merit of the idea, rockets that can take off and then land vertically without damage were also laughable pre-SpaceX.

[altcognito]

NASA had literally done it. It was never laughable, just thought to be an incredibly difficult engineering challenge.

I think compute in space suffers less from being "impossible" and more from being "impractical". It is plenty easy to put compute in space. It is just still silly expensive and by the time your equipment makes up the cost of putting it in space, it will be well out of date.

[selectodude]

The non-deterministic nature of LLMs will get a whole lot more fun when we add randomly flipping bits to the situation.

[jrflo]

Space compute isn't real though, it's just a scheme to pump the value of SpaceX before IPO by associating it with AI. It's really hard to cool things in space, because there's no matter to transfer the heat away. All you've got is radiative cooling, and that's really really slow.

[skywhopper]

He believes in the value of the idea of “space compute” for attracting investors to SpaceX. But the existence of the idea of “space compute” as a better way to deploy datacenters (along with everything else Musk has claimed in the past decade) should give everyone pause as to the plausibility of literally everything else he says.

[oblio]

> should give everyone pause as to the plausibility of literally everything else he says.

That ship has sailed, we're in the Age of Cults. If you're a believer you're probably also invested in his companies, so your mind is doubly-clouded.

[geodel]

> so your mind is doubly-clouded

Further I think external agents are controlling their mind.

[jpadkins]

> along with everything else Musk has claimed in the past decade

* mass produced EVs * Neural links to brain * reusable rockets * more efficient tunnel digging

Dude has a pretty good record of taking stuff in R&D and making pretty real products and/or companies. Can you name someone better?

[ctoa]

Mass produced EVs and reusable rockets are already decade old accomplishments. His track record this last decade of accomplishing things he proposes is noticeably much worse.

Neuralink is just another medical device company right now. The things it is succeeding at aren't the crazy sci-fi Elon "telepathy" vision at all, and there are other BCI companies.

Boring Company is quite stagnant and I would say what they achieved is very underwhelming. The original framing was that it would revolutionize the costs of subway and high speed rail tunnels and it has not achieved that at all. The things it currently does are nothing like the vision Musk was selling in 2017, much of the original concept was shelved.

Robotaxi is wildly off schedule and his timelines that he puts out for it have been non-stop over optimistic and wrong. This may catch up eventually, but it's currently not at the point that Musk was already claiming it was at 10 years ago.

I'd list as another big flop of the last 10 years the Tesla Solar Roof.

Also, I think by far the biggest Musk reality mismatch is the story he pushes regarding Optimus vs what anyone has actually seen of that product. That mismatch is absolutely wild. Anyone who believes the things he says about timelines for it needs to get a grip.

[pyrale]

> He believes in [...]

...the value of having others buy the idea of space computing. I don't think he himself believes in what he says.

[jameson]

Right, also Anthropic has been having difficult time getting more GPUs

[giancarlostoro]

I think its a mix of both of your first two points. He has hinted that he will be producing chips for Tesla and SpaceX here in the US in the future, so there will be a time where Elon will have every critical component we need for Tesla and SpaceX here in the US and not overseas, and fully in-house. Once he does this, nothing stops him from selling components. He's also insanely efficient at standing up data centers, so I would not be surprised if he gets enlisted by someone like Anthropic to setup infrastructure for him.

[notfromhere]

xAI is a dead company; you don't sell compute if you're growing.

More promising is that cursor is training a model using it.

[mitjam]

Interesting that nobody mentioned SAR in this thread, yet.

SAR benefits from space compute by reducing the strain on downlinks (pre-process, detect in space). A large-ish (think 6k sat) constellation of much bigger sats (2-2,5 ton) than current Starlink sats can run inference on board, distribute load to nearby sats, and enable full round-the-clock surveillance of the whole world.

The nominal compute capacity is in ballpark range of a modern AI data center, but it's only about 20% used on average due to duty cycles. An indestructible, global eye.

The big Starship launch vehicle is perfect platform to bring them into orbit, can maybe bring 40 at a time into orbit, so 150 launches for 6k. Maybe even fewer sats are needed, depends mostly on electrical efficiency of the components.

[Glohrischi]

That would be absolutly ludicris.

His own product is competing against anthropic.

[cbm-vic-20]

It's a money maker. Maybe not a number one spot, but they don't have to stand up any more datacenters.

[Glohrischi]

His product should be the money maker.

[fillskills]

We could see the first company vertically integrated from etching to chip to data center

[AlotOfReading]

That's Intel. Probably IBM too, though they've been doing mass manufacturing with TSMC (and GloFo before) for years instead of their fabs. I wouldn't be surprised if HP did similar things back in the 80s.

[jacobrast]

What do you mean by etching? Google does also it's own chip design with TPUs, data centers, and models but afaik only TSMC Intel and Samsung do the actual semiconductor fabrication

[ericd]

Assume he's referring to SpaceX's https://en.wikipedia.org/wiki/Terafab

[oblio]

> Analysts estimate the costs for the full-scale facility at between US$5−13 trillion.

https://en.wikipedia.org/wiki/The_Line,_Saudi_Arabia

Musk and MBS should hold hands.

[gpm]

Turning The Line into one giant ~~paperclip~~ chip assembly line!

We can call it the Peta(Principle)Line!

[Glohrischi]

We wont for anytime soon.

Intel and TSMC are not what they are today just because they buy a very expensive EUV Machine from ASML but because they have the knowledge and infrastructure to even use these machines.

[sofixa]

IBM was doing this since the 50s.