Elon Musk has been predicting a new era of AI for more than a year. SpaceX’s CEO said it would be “easy” to build an AI data center in space and that it would become the cheapest place to deploy AI within two to three years. However, the company’s pre-IPO filing shows a much more conservative outlook.
SpaceX is gearing up for what could be the largest initial public offering in history, with a goal of raising $75 billion and valuing the company at about $1.75 trillion. The U.S. Securities and Exchange Commission (SEC) requires companies to file an S-1 statement before going public as part of informing potential investors of the risks. SpaceX’s S-1 application, reviewed by Reuters, reportedly acknowledges that an orbital data center may not be commercially viable.
“Our efforts to develop in-orbit AI computing and in-orbit, lunar, and interplanetary industrialization are in the early stages, involve significant technical complexity and unproven technologies, and may not achieve commercial feasibility,” the filing states, according to Reuters.
Gizmodo was unable to view or confirm the document, and SpaceX did not respond to a request for comment by the time of publication.
reality check
This cautious application is a far cry from the lofty ambitions SpaceX outlined in its FCC filing in late January. The company requested permission to launch an in-orbit data center constellation of up to 1 million Starlink satellites, arguing that taking advantage of the “near-constant” solar power available in orbit would reduce operating costs, energy demands, and environmental impacts associated with ground-based data centers.
“The launch of a constellation of one million satellites to act as orbital data centers is the first step toward becoming a Kardashev II civilization, a civilization capable of harnessing the full power of the sun, while supporting AI-driven applications for billions of people today and securing humanity’s multiplanetary future among the stars,” the application states, according to SpaceNews.
Of course, the primary purpose of the S-1 is to disclose risks to investors, so it is not surprising that a more cautious tone emerges in pre-IPO filings. Beyond commercial feasibility issues, SpaceX acknowledges significant technical hurdles, warning investors that future orbital data centers will operate in the harsh and unpredictable environment of space and will be exposed to “a wide range of unique space-related risks that can cause malfunctions and failures.”
Indeed, scientists, satellite experts, and SpaceX’s competitors have publicly criticized the plans outlined in the FCC filing, arguing that current technology and capabilities are insufficient to build and operate an orbiting data center, much less a constellation of 1 million. For SpaceX, part of the problem is that the satellites and the rockets used to launch them aren’t yet ready.
Musk said SpaceX could build an in-orbit data center by “simply scaling up its Starlink V3 satellites,” but the company has yet to debut it. SpaceX plans to launch using its own Starship rocket, but the full high-speed reusability and launch cadence needed to build an orbital data center has not yet been demonstrated.
“Any failure or delay in the extensive development of Starship or in achieving the required launch cadence, reusability, or capabilities would delay or limit our ability to execute on our growth strategy,” the S-1 filing states.
It’s refreshing to see SpaceX finally acknowledge the huge challenges and risks standing in the way of its orbital data center dream. While this may scare off some prospective investors, at the end of the day, SpaceX’s long-standing dominance in the commercial launch industry and desire to tap into the burgeoning AI market still make it an attractive investment.
Whether the risks outweigh the potential benefits remains to be seen.
#trillion #flipflop #SpaceX #orbital #data #center #profitable