Starcloud raises $250 million for orbital data centers as launch options dry up
Starcloud, a startup building satellites that perform AI inference in orbit, has raised a $250 million extension to its Series A funding round, valuing the company at $2.3 billion. The move comes as CEO Philip Johnston seeks to secure launch capacity amid growing constraints in the rocket market, with SpaceX's workhorse Falcon 9 due to be phased out by 2028 in favour of the larger, still-unproven Starship, while rival vehicles from Blue Origin, ULA and Rocket Lab are not yet flying regularly.
The funding will let Starcloud open a bigger manufacturing facility and advance its largest spacecraft, Starcloud-3, which is designed to launch on Starship. The round was led by Manhattan West Ventures with participation from Nvidia (reportedly contributing $25 million) and Cisco, alongside Benchmark, EQT, Soma, NFX, 776, Cedar Capital, Goanna Capital and Standard Capital. Starcloud has requested FCC approval to operate 88,000 spacecraft, plans to launch its next-generation Starcloud-2 satellites on rideshare flights in 2027, and is notable as the only known operator of an Nvidia H100 GPU in orbit.
- Starcloud raises $250m, now valued at $2.3 billion
- Funding secures manufacturing and future launch capacity
- Nvidia, Cisco among investors as rocket launch options tighten
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Advocates argue that orbital data centres address real, worsening constraints on Earth: land, power grid capacity and cooling water needed for AI computation are all scarce and contested, whereas satellites in the right orbit can draw on near-continuous solar power and the vacuum of space for passive cooling. From this view, securing launch capacity now and building manufacturing scale is a prudent, forward-looking bet by sophisticated investors including a major chipmaker, positioning the company to serve an AI compute boom that terrestrial infrastructure may struggle to meet, and the willingness of investors like Nvidia and Cisco to back the round signals credible technical and commercial validation rather than blind speculation.
The case against
Sceptics contend that the venture rests on unproven foundations: it depends on Starship, a rocket that has not yet demonstrated routine commercial reliability, while established alternatives are being phased out, creating real risk that the launch capacity underpinning the business plan simply will not materialise on schedule. They would also point to the $2.3 billion valuation and the request to operate 88,000 spacecraft as reflecting speculative enthusiasm for an AI-driven narrative rather than demonstrated economics, noting that orbital data centres face unresolved challenges around radiation-hardened hardware longevity, data downlink bandwidth and servicing costs that could make the concept far costlier in practice than earthbound alternatives it aims to replace.