Comment by small_model
2 hours ago
Same with data centres in space, "Noooooo it cant work, heating in space, blah blah" same people, will never achieve anything in life as they only see limits.
2 hours ago
Same with data centres in space, "Noooooo it cant work, heating in space, blah blah" same people, will never achieve anything in life as they only see limits.
Different category of doubts. It's unreasonable to say starship can't work in a world where the Saturn V did work.
For datacenters in space to work, they have to solve a bunch of unsolved problems and be more economical than datacenters in earth.
To be fair, radiators in space doesn't require any new technology. It's a very well understood problem. The best method is infrared radiators. Cooling is just a function of how large the surface area of the radiators, their heat, and the cooling required.
Where I think you're correct is that typical AI data centers require a LOT of cooling, and to achieve this at scale would require enormous radiators. So large that it's hard to square the business case given current launch costs. I've read a lot of research in this space and it looks like the much more realistic use cases (at least at first) will be sub-processing. Meaning processing data already in space. These could be telemetry, Starlink, tracking, telescopes, logistics, military, GPS, weather, deep space missions, first-response, alerts, etc. In particular, applications which require lower latency.
There are also other applications which are less cost sensitive. For example, applications which might be banned on Earth, or at risk of espionage, attack, or intrusion.
If we want to make typical AI data centers in space to be economical, we need launch costs to drop to under $200/kg. Interestingly, [Starship could reduce costs down to $67/kg.] Even lower with >9 launch cycles. (https://arstechnica.com/space/2026/07/rocket-developers-used...) This would make the business case *extremely* attractive.
Great comment and I agree with all of it except you're describing an expanded starlink rather than "datacenters in space". There are definitely workloads that benefit from being in space but that's different from being an economical place to put a rack of H200s doing training or inference.
Dissipating that much heat in the first place would be a first, you also need a butt ton of power so maybe you can shade your giant radiators with giant solar panels, but all of this needs to unfold from something launch-packaged.... AND be cheaper than just plugging them in on earth.
1 reply →
funny that skeptics didn't pile on Google for the same plans... https://www.nytimes.com/2026/09/24/technology/google-suncatc...
"It's unreasonable to say starship can't work in a world where the Saturn V did work."
Most of the doubts were economical as well. Saturn V was extremely expensive, its per-unit cost could be expressed in still-human-readable fractions of the entire US GDP. Starship should be several orders of magnitude cheaper while providing more capabilities.
I don't doubt Starship, BTW. Great project.
Totally, just like all the haters of the Hyperloop!
You’re right, I’m just back from the city, took no time and was so cheap thanks to the hyperloop. Next week I go to Asia, I think I will use the rocket transport system Musk announced a few years ago. It’s so nice how my Tesla is making me so much money while I’m traveling, offering taxi service to everybody around. Next will I will get the add-on to have it join the Tesla compute fleet - it’s free compute if you think about it, given the solar roof is generating the energy! Might even get one of those humanoid robot to replace me at work. Just hope it’s not raining too much, my cybertruck steel plates are rusting a bit. Also, my kids are really enjoying their summer camp at the Mars colony!
Limits are security. Dont have to do this, don't have to risk that, no need for uncertainty, anxiety won't kill the weak confused mind...you can't fail if you never try (big brain!). That's the type. It is called cowards ;)