SpaceX’s IPO filing this week wasn’t about rockets. Or Starlink. Wall Street already understands these businesses. Launches are booming. Satellite internet is expanding profitably. These stories are mature enough to be modeled in a spreadsheet.
The real pitch, which justifies increased valuations over the next decade, is something much stranger: AI data centers in orbit.
In its application, SpaceX describes plans to deploy “AI computing satellites” into sun-synchronous orbit starting as early as 2028. The idea is simple in theory, but preposterous in practice. Moving AI infrastructure to extraterrestrial space. There, solar power generation is virtually unlimited and cooling occurs naturally through radiant heat dissipation.
The AI race is increasingly about who can control the computing and generate tokens most efficiently. SpaceX claims that its current bottlenecks are physical, such as power generation, data center construction, and chip manufacturing. The company believes that ground infrastructure will not be able to keep up with the exploding demand for AI, especially as inference models and AI agents consume exponentially more compute.
Hence the orbital pivot.
Actually, the logic is pretty consistent. In orbit, solar arrays receive nearly constant sunlight. There is no atmosphere, no weather, no NIMBY to prevent us from permitting another gigawatt data center. SpaceX says space-based solar arrays can generate more than five times the energy per unit area of ground-based systems. Combining Starship launches with Starlink networking, Elon Musk sees a future where computing itself becomes the new space industry.
And here’s the important part. This means Musk isn’t just freestyling another sci-fi fantasy. Google is pursuing the same idea.
Late last year, Google launched Project Suncatcher and published a paper proposing a “space-based ML data center” using solar satellites networked with optical links. This paper argues that the future energy demands of AI may require computing infrastructure to take off.
Google is already testing the radiation resistance of its TPU chips in simulated space environments and working with satellite specialist Planet Labs, which went public in 2021, to design AI satellites. The search giant is also reportedly in talks with SpaceX about launching future AI satellites into orbit.
Most importantly, Sundar Pichai publicly supported the concept.
“There’s no doubt that in about 10 years you’ll start to see that as a more normal way to build data centers,” he told Fox News.
That quote is important. Pichai is perhaps the least theatrical CEO of a major technology company. Even if Musk and Pichai independently conclude that AI computing will eventually take off, investors won’t be able to dismiss the idea as pure science fiction.
Significant risks remain. SpaceX acknowledges that the technology may not be commercially viable. The required firing cadence is incredible. The engineering challenges are tough.
But IPOs are about faith in the future and the promise of near-monopoly in technology.
And SpaceX just told investors that the monopoly it wants to own is AI infrastructure in space.
Sign up for BI’s Tech Memo newsletter here. Please contact us by email. abarr@businessinsider.com.