The idea of putting data centres in orbit sounds like the kind of thing that gets announced at a conference and quietly shelved. It is not.
Multiple companies are now pursuing orbital computing infrastructure in earnest, and the money behind the push is serious enough to suggest this is forming into a genuine industry rather than a collection of moonshots.
SpaceX is the most significant name in the mix. The company has filed confidentially for an IPO targeting a $1.75 trillion valuation and a $75 billion raise, and the prospect of orbital data centres sits neatly inside that story.
SpaceX's primary business is as a launch company, with Starlink providing a second revenue stream.
Data centres in space would add a third, and crucially, SpaceX would also book the launch revenue for putting them there. For a company preparing to go public, that kind of vertical integration is a compelling pitch to investors.
The startup making it real
The more immediate story may be StarCloud, a Y Combinator-backed startup that has raised $170 million and achieved unicorn status while most people were not paying attention. StarCloud is among the first companies building a business specifically around orbital data infrastructure, and its funding round puts it well beyond the concept stage.
Blue Origin, Jeff Bezos's space company, is also working on its own satellite network and computing capabilities in orbit, setting up what looks like a three-way competition between the Musk, Bezos, and venture-backed camps for position in a market that does not yet fully exist.
Why now
Part of the answer is straightforwardly practical. Building data centres on Earth has become harder. Planning opposition, energy constraints, and local resistance have slowed rollouts in several markets. Orbit has none of those problems, at least not yet.
There is also a financial logic specific to this moment. Companies approaching public markets need growth stories, and the data centre boom driven by AI demand has made compute infrastructure one of the most-watched sectors in technology. Orbital data centres give SpaceX in particular, a way to extend that narrative beyond rockets and satellite internet.
The sceptic's case
The honest assessment is that orbital data centres remain small additions to overall compute capacity rather than any kind of replacement for what is being built on Earth. The engineering, physics, and orbital mechanics involved are genuinely complex, and even optimistic projections put the contribution to global compute at the margins for years to come.
There is also a scenario where slowing demand growth or a reassessment of AI infrastructure requirements takes momentum out of the sector before orbital projects reach commercial scale, leaving early movers exposed.
For now, though, the money is moving and the competition is forming. That combination tends to make things happen faster than sceptics expect.