Skip to content
XAI's Radical Plan: Data Centers In Space episode artwork

Feb 12, 2026 18 min

XAI's Radical Plan: Data Centers In Space

from The Daily AI

In this episode, we unpack Elon Musk's ambitious vision for XAI, including its new organizational structure and a detailed roadmap to establish AI data centers in orbit. We explore the strategic rationale behind this move, the technical challenges involved, and how it ties into the broader SpaceX and X ecosystem. Chapters 01:51 XAI's New Structure and Teams 08:10 X's Growth and AI Adoption 10:12 Orbital AI Infrastructure 15:35 Starship's Role & Cost Reduction 18:36 Technical Challenges & Competition Links Get the top 40+ AI Models for $20 at AI Box: ⁠⁠https://aibox.ai AI Chat YouTube Channel: https://www.youtube.com/@JaedenSchafer Join my AI Hustle Community: https://www.skool.com/aihustle

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

XAI's Radical Plan: Data Centers In Space

0:00 18:45
of matches

Transcript Auto-generated, may contain errors

Welcome to the podcast. I'm your host, Jayden Shafer. Today on the show, I'm talking about something pretty wild that just happened. Elon Musk just hosted a 45 minute All Hands meeting on X, and he did it for XAI and basically laid out their entire roadmap for getting data centers in space.

This is something, I've seen a bunch of articles that are kind of critical. They don't think this is possible. A lot of people say this is just kind of an IPO play. And I think Elon's really making the case that he is going to try to do this.

I think what's really interesting with this in particular is that this was an All Hands meeting for the company and it publicly posted. I think this is, I mean, you wouldn't see Boeing or any of these other competitors doing this. Usually everything has these road maps and there's all these NDAs and there's these really carefully scripted keynotes. To me, this was really interesting to just see, the whole thing laid out and the whole plan just put out to the public.

So in the video, I'm going to break down basically what Elon Musk's long-term vision is for XAI and also the near-term product execution. We'll talk about some of the infrastructure strategy and how X and SpaceX and XAI are all kind of tying together. This is a fascinating conversation. Before we get into all of that, I wanted to say, if you want to try any of the top AI models, including all the latest from XAI, like their Grog model and everything else, they're rolling out.

I'd love for you to try out AIbox.ai, my own startup. I let you access over 50 of the top AI models, image, audio, text, and we just dropped a brand new tier, which is 899. So for 899, you get access over 50 of the top AI models. Basically, you consolidate all of your different subscriptions into one place, you chop them all in the same place and you don't have to go shuffling through tabs or log into different accounts to try to find where you had a conversation or if you want to try the latest from anthropic or Google Gemini or whatever, it's all in one place for 899.

So if you want to try it out, it's AIbox.ai. You also can vibe code, really cool tools and post them. So go check it out, I'll leave a link in the description. I think a lot of the address was kind of focused on a whole bunch of people that have recently just left XAI.

There was a bunch of the original team members, a lot of founders of XAI that left. This is what he said about it. He said, as the company grows, especially as quickly as XAI, the structure must evolve. And this was actually an exposy put out.

He said, this unfortunately required parting ways with some people. I think reorganizing the side of these companies are growing very fast is quite common. I think what was interesting to me with all of this is kind of the new structure. So XAI is now splitting into four primary teams.

They have Grock, which is the voice, you know, the LLM and the voice that goes with it. And basically all the core chatbot experiences. They then have the coding system, powered software generation team. So that's really kind of being like a cloud competitor is what they're trying to build.

And they actually do have a model that ranks quite well on the benchmarks for software development. It might even be number one at times. But because they don't have all the integrations and custom stuff, the cloud code has cloud code as still far as usage goes as way ahead. Then they have the Imagine.

This is the video and image generation components of XAI. And then they have Macrohard. This is, I mean, basically, I think they're just trying to make fun of Microsoft and call it Macrohard. But basically it's an AI initiative, which they're kind of focusing on full computer task automation and then corporate scale modeling.

And really just trying to compete with everything that Microsoft makes and just call it Macrohard. So anyways, it's classic Elon. Macrohard is going to be led by Toby Flynn. And that is, I think, maybe one of the most, I guess it's just a really big play.

The goal is that they're not just trying to assist with workflows. They're actually trying to automate them. So yeah, in addition to perhaps making some of the software that Microsoft has and embedding AI into it, I think more likely what they're going to be doing is doing complete automation software and tools for enterprises. It's obviously going to be a very hot thing.

We see just with like, and it ends up here in a lot of these companies that are really crushing it there. And more likely, you see the comet perplexity kind of tools or just basically all these agents that anthropic and opening our building. I think that's where they're going to be going to be more of the agents. So Toby who's leading this all up was talking about it.

And he said that Macrohard is able to do anything on a computer that a computer is able to do. There should be rocket engines fully designed by AI. So that is pretty crazy. I think if you look at that, obviously that makes it really clear that like, XAI isn't just kind of trying to chase chatbot improvements on Grog.

They're also targeting a lot of general purpose, kind of high leverage intelligence systems. And they're also trying to get into software like operating on a lot of these agents that can work on software. And personally, the best tools I've seen for this so far have come out of anthropic. It's been the one that I've been most impressed with.

But they do have, there are a number of other players, OpenAI and Microsoft and perplexity that are kind of all chasing this as well. So as far as growth is going right now for the whole company, X and XAI because XAI acquired X before SpaceX acquired XAI. So it's hard to keep mind around all of these acquisitions. I'll have the letter X in them.

But in any case, talking about X specifically formerly Twitter, Nikita Bear, he's had a product. And he said that they have just crossed a billion dollars in annual recurring subscription revenue. I think this was a whole bunch of this was driven by holiday marketing pushes and then a lot of product expansion. There's a bunch of executives over there that said that imagine their image generator and video generator is generating tens of millions of videos per day and billions of images over last month.

And that's according to my internal metrics. That is a lot of usage. I think it shows and kind of reflects a bit of a broader thing, which is basically AI is no longer experimental, right? It's pretty mainstream.

And we're seeing this with JAD GPT and a lot of these other players. But it is being pushed out on every platform everywhere. And a lot of people might call this AI slop and especially with Gurok and their image and video generation tools. I think there's a lot of problematic things there.

But it's definitely hit scale. Like there is billions of billions of images being generated. So a lot of those presentations were just kind of what's going on internally at the company. The most interesting one that everybody is talking about is when the whole presentation they were giving kind of came towards the end, Elon started talking about what he's been talking about forever, which is space space AI infrastructure.

So he kind of just reiterated what he has said in the past that the cheapest place to run large scale AI within a few years is going to be an orbit. Basically the benefits of the space has constant solar energy, fewer land permitting bottlenecks. Although, I mean, you've got to still get your satellites approved and stuff, but apparently that's easier than land permitting on Earth. And then there's also path to scale compute beyond just terrestrial constraints.

So SpaceX has already requested regular theory of approval to build solar powered orbital data centers. I think we've seen that the administration looks like is kind of looking at this favorably and is going to grant this. I think they put out a tweet recently where they were saying they're requesting comments. If anybody has any objections, speak now or forever hold your piece basically.

And I think this is going to have their application that they put out for was like for a million orbital data centers. So there's going to be a massive satellite constellation. Elon has floated the idea. And I mean, a lot of people think this is like science fiction, but I think there's a lot of real engineering that goes into this.

So one of the elements that a lot of people are talking about is lunar manufacturing. So literally building like factors on the moon that can launch and develop and create these data centers. And I think one of the parts of this is as far as regulation goes, you have to I think de-orbit your satellites every five years or have a plan for de-orbiting so that we don't just have unlimited space junkings after you'll basically come down. A lot of people have it where they just re-entilate and fear and burn up.

And that's kind of their de-orbiting plan. And it feels like we might have perhaps ways to capture them and bring them to the moon to get retrofitted. Or perhaps we're still going to let them just burn up on re-entry and just keep our factor on the moon pumping out more and more solar panel covered space compute data centers. It's a really interesting thought here.

But in any case, I think one thing that was said about Elon, I was interested in. He said, it's difficult to imagine what an intelligence of that scale would think about. Perhaps it's going to be incredibly exciting to see it happen. I think there are a lot of critics that are kind of focusing on some of the cost comparisons between ground and orbital data centers.

I think they often are kind of assuming static launch and economics, which is basically entire premises hinging on Starship here. If Starship is able to achieve its intended cost reductions per kilogram to orbit, then I think the business case is going to shift dramatically. But right now, because it is so expensive, even for Starship to get things into orbit, this doesn't feel like a profitable business. He's like, it's way cheaper because there's like no land permitting.

And we don't have to get water to cool these things. And we have unlimited solar. And so those things are true. But just getting these satellites into space is expensive.

But Starship has a plan to decrease their payload costs. So being able to get things into space for cheaper, they're planning on doing a lot of optimizations. And if they are able to do those optimizations, perhaps this becomes a profitable business. Today, the Falcon 9 already delivers basically the best industry pricing for getting stuff into space.

Starship is trying to push that down by an order of magnitude. There are no other launch vehicles in development. And right now, they're targeting comparable reusability at that scale. So part of basically saving a lot of their costs is how much of the rockets can be reused if they shoot something up and they bring it back down.

And that's the thing that they save it by a killer of money over NASA was just that the rockets now are not exploding or crashing on reentry. They're actually able to reuse them. So how many times can they reuse them? How many parts can be reused?

How much they need to be repaired? I think you can bring all those costs down and just have a vehicle that goes up and down to space over and over again. That's where they start saving a lot of money. And they have a goal and a plan for that.

So like theoretically, it's possible. But it's not a reality today. I think there's a lot of real challenges. There's radiation hardening.

There's thermal management. There's satellite manufacturing costs. There's intersatellite communications throughput. SpaceX has already shown that they can obviously mass produce and operate thousands of satellites.

They have the Starlink constellation, which by the way, if you didn't know, Starlink generates 80% of SpaceX's revenue. They charge money to shoot things into space. And that's kind of what their company was built on. But Starlink is actually the biggest revenue generator for their company right now.

So obviously they're able to do it. They can mass produce. They can mass operate thousands of satellites. And they're doing a good job of that.

And so turning those satellites into data center. So it's kind of the next step of can they do it. And if any company can, it would appear, they have the most experience they would be able to. But it's tricky.

And I think the one thing that's going for them, there's not a ton of competition. That is obviously some insane operational experience. And that's a huge moat. Not a lot of people are really competing directly.

But I will say Google's project Suncatcher is focusing on. There's a startup called StarCloud. And then even Amazon and Blue Origin are kind of showing that this orbital AI infrastructure is something that they're interested in. So there are other players that are competing.

Google, Amazon, and StarCloud. And so, and then of course, SpaceX. So we have like four players in this. And to me, that's exciting.

It feels less fringe and more like there's kind of a race. And so we'll see which of these players can execute fastest and best. And I love the competition and all this. A couple of things that I wanted to say from like a technical standpoint.

And some of the things that are tricky in all of this. So basically, early orbital deployments are more likely to focus on inference rather than full scale training. So these data centers out there, they're not going to be doing full scale training. They're probably going to be doing inference.

Which is just when I'm chatting with talking to chat GPT, it has to run inference to basically give me my answer. And that's what they're probably going to do. So inference workloads, they can run on smaller GPU clusters. And they can tolerate distributed architectures a lot more easily.

And so there's not like one giant data center in space. It's going to be a ton of these little tiny satellite data centers all linked together. And because they're distributed like that, you have to have unique architectures that can actually make that function. So basically, that opens up some near-term paths.

There's like, you can imagine customer service agents, general media workloads, AI-assisted software tasks, which could be computed in orbit and then powered by continuous solar panel. That kind of stuff all makes sense. But at the end of the day, as one engineer put it, a flop is a flop. And it doesn't matter where it lives.

So if compute is fungible across terrestrial and orbital environments, companies like SpaceX and XCI can dynamically allocate where it's cheapest and most scalable. If they have the capacity up there in space, basically, then wherever, if their servers are getting overwhelmed, their data centers are getting overwhelmed on Earth, even if perhaps it's a little bit more expensive in space, let's say, but they're overwhelmed here. They can shoot it up and use some of that compute power in space. So they can work their system that way.

And alternatively, they would just have you going with the competitor, which would be upselling, upcharging them a lot, AWS or Google. And so they would have to add their own margins and markups. So it probably is comparable, except they're not giving money to their competitors perhaps. And they're able to scale up faster.

And hopefully bring the cost down in the future. I think the advantage of this kind of vertical integration that they might be doing is kind of obvious. SpaceX obviously can control the launches. They operate all the satellites.

And they now own XCI. So I think with kind of having this huge combination, one has a really unusual ability, basically. I think I don't know how to describe this to roll out some crazy experiments. He has the full stack.

He's got energy, compute, application. Everything needed to pull this off. He has the pieces. It's obviously a big mission and a big opportunity.

But I think he would be able to pull this off. So there's a recent co-founder who just left XCI. And there's all these headlines about it. But I think if you zoom out, XCI has over 1,000 people.

And they're continuing to hire very aggressively in Frontier AI, like OpenAI and anthropic and Google Gemini, all these big labs. I think it's really common for the early stage builders to move on once the company kind of transitions from a scrappy lab, which XCI has been for a long time. And now they're kind of scaling up and turning to an execution machine. It's more of a grind.

There's deadlines. We saw this with OpenAI. I mean, basically, the entire founding team has left and started their own projects. I think we'll see the same thing with XCI.

I think a lot of the departing engineers that are leaving, as I said, they're excited to start new projects, which is exactly what happened with OpenAI. So I don't know. A lot of people are making big deals. Like, oh my gosh, everyone's leaving XAI.

But it's like, yeah, well, that happens to OpenAI. And everyone else, people just leave to start their own things. I think it's basically the signs that the ecosystems help me. There's more money to be made.

There's more projects to be worked on. More new ideas because people are talented. And I'm all for it. Just make more competition, make more innovation.

Let's get the talent flywheel moving. So I'm excited about that. What the long-term play is for this company that they're building right now? I think Elon, for many years, has talked about AI not just as a product category, but as a civilizational force multiplier.

I think right now, we're seeing a lot of those pieces aligned. There's reusable heavy lip rockets. There's massive satellite manufacturing capacity. There's vertical integrated AI research.

And then, of course, he's got a social platform for distribution, but also for data. And he's got data. He's got AI. He's got manufacturing.

He's got so many pieces. They are and have all of the pieces to this. So it's obviously a massive vision, but it's also kind of consistent with his experience and what his other companies are able to produce. He kind of merges them together.

And Earth is already very, this is massive for productivity. So perhaps putting it in space is going to remove the energy and land constraint that's limiting scaling. I mean, those are two of the biggest scale limiters. I've talked to many investors focusing on this kind of stuff.

And the biggest concern is there's not enough energy for these data centers being built. And then it's going to have a really negative impact. And they're like, we got to do nuclear. But nuclear takes a lot of regulation.

And building these nuclear facilities takes a long time. So this might be an answer. Definitely some massive engineering problems and challenges like this isn't going to be easy. But SpaceX has an awesome track record of doing what was basically seemed impossible.

If you've told people in SpaceX who started there and have reusable rockets that went into space or into orbit and then came back down and were able to be reused, NASA would have, I mean, in fact, go look at all of the most famous space explorers, engineers. When Elon was starting, they all laughed at him. And I mean, he was deeply offended, had to show him. I mean, but that's probably a good thing, right?

It's always good to feel like you have to prove something because he did. He showed everyone that he could make something incredible. So I think whether or not a million satellite data centers arrive on the original timeline, he's usually optimistic with timelines. So whether he gets all a million up in time, it doesn't really matter.

I think the direction that the industry is going is pretty clear and he's now kind of pulled in Amazon and Google and some other big competitors to compete there. And I think compute is becoming basically the core infrastructure right now of the 21st century. Elon is betting that the next step in that and the next frontier is going to be at the edge of the atmosphere. So it's gonna be exciting to see.

I'll definitely keep you guys up to date on everything getting developed in this space race, the new space race. Thanks so much for tuning into the podcast. If you enjoyed the episode, if you learned anything new, I would super appreciate it if you could leave a review wherever you get your podcast on Apple or Spotify. It helps to show out immensely and it makes my heart happy.

So if you do that, I'd appreciate it. Make sure to go check out AIBox.ai and check out our brand new 899 subscription tier to access to over 50 of the top models. I mean, you basically get everything that you want out there from the top labs and for it's 899 a month. And yeah, it's amazing.

So go check out, leave a link in the description, AIBox.ai. Catch you guys in the next episode.

Frequently asked questions

How long is this episode of The Daily AI?

This episode is 18 minutes long.

When was this The Daily AI episode published?

This episode was published on February 12, 2026.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this The Daily AI episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.