Sam Altman and Elon Musk exchanged pointed criticism on social media over the weekend, drawing renewed attention to questions surrounding the viability of space-based data centers as a near-term business. After Musk accused Altman of being a scammer, Altman responded by suggesting Musk was overselling short-term prospects for space data centers to public market investors.
The exchange highlights a broader gap between investor enthusiasm and expert skepticism regarding SpaceX’s plans to deploy orbital data centers for AI inference. The initiative has been cited as a key driver behind SpaceX’s roughly $2 trillion valuation, with bullish analysts pointing to the potential of orbital compute to support SpaceXAI’s models or function as a space-based cloud provider. However, industry experts, including founders of competing space data center startups and engineers who have studied the economics, have consistently concluded that the business case remains unworkable until launch costs fall dramatically and high-powered satellites can be manufactured cheaply at scale.
Much of SpaceX’s case rests on Starship, its next-generation rocket, which is expected to attempt its 13th test flight as soon as July 16. Even if the vehicle successfully recovers both stages, achieving reliable, repeatable reusability is expected to take years, with Starship’s near-term launch cadence likely prioritized for NASA missions and SpaceX’s Starlink network rather than data center satellites. SpaceX has also acknowledged during its IPO process that Starship may not achieve full reusability in the near term, potentially requiring the vehicle to discard its second stage on each flight, undermining the economics of orbital data centers.
While Musk maintained that SpaceX could begin flying data center satellites next year, analysts noted that launching a single satellite differs significantly from establishing the manufacturing and launch scale needed for a viable business, a milestone many experts believe remains unlikely before the 2030s.
