Google SpaceX Orbital Data Centers Talks
Google SpaceX orbital data centers talks reported May 12, 2026; traders weigh effects on launch demand, SpaceX pre-IPO strength and orbital launch costs.

KEY TAKEAWAYS
- WSJ reported Google and SpaceX were in early talks for a rocket-launch deal for orbital data centers.
- SpaceX had filed with the FCC for authorization of up to 1,000,000 orbital data-center satellites.
- No deal terms or timelines were disclosed and Google had engaged other launch providers.
HIGH POTENTIAL TRADES SENT DIRECTLY TO YOUR INBOX
Add your email to receive our free daily newsletter. No spam, unsubscribe anytime.
Google and SpaceX entered early talks on May 12, 2026, for a rocket-launch agreement to support orbital data centers, potentially boosting launch demand and SpaceX’s business ahead of its public listing.
Talks, Corporate Links, and Regulatory Filings
Google’s Project Suncatcher aims to deploy prototype satellites by early 2027. The design features tensor-processing units (TPUs) operating in a roughly 650-kilometer dawn-dusk orbit, arranged in clusters of 81 satellites. The project was announced in November 2025.
The companies have existing financial and operational ties. Google disclosed a $900 million investment in SpaceX in 2015 regulatory filings. SpaceX expanded its computing reach after acquiring xAI in February 2026 and recently arranged computing capacity from an xAI Memphis data center to support Anthropic.
On January 30, 2026, SpaceX filed an application with the Federal Communications Commission (FCC) seeking authorization for up to 1 million orbital data-center satellites in orbits ranging from 500 to 2,000 kilometers. The FCC Space Bureau accepted the filing on February 4, 2026. The application provides limited public detail on satellite mass, cost, and schedule but has raised concerns about collision risk, disposal, and spectrum management amid roughly 14,500 active low-Earth-orbit satellites.
Technical and Economic Challenges
Economic viability remains a significant hurdle. Google’s Suncatcher modeling suggests orbital data centers could be feasible if launch costs fall to about $200 per kilogram by the mid-2030s. A private estimate from Starcloud sets a cost-competitive threshold near $500 per kilogram, assuming SpaceX’s Starship achieves frequent launches in 2028–2029. Current launch costs are near $7,000 per kilogram, implying a roughly 35-fold reduction is necessary.
Engineering challenges are substantial. Proposed inter-satellite links would require speeds around 10 terabits per second, compared with current systems delivering 1 to 100 gigabits per second. Radiator mass to dissipate heat from about 10 megawatts of on-orbit computing is estimated at roughly 529 metric tons. Designers note that routine human maintenance would not be possible.
Smaller commercial pilots have advanced. Starcloud launched an Nvidia H100 GPU satellite in November 2025 and ran a trained large-language model on orbit in December 2025. The company raised $170 million in a Series A round at a $1.1 billion valuation in March 2026. A Thales Alenia Space ASCEND study projects a minimum viable orbital capability near 10 megawatts by 2030 and about 1 gigawatt by 2050.
Executives and investors offer mixed views. SpaceX told prospective public-market investors the orbital data-center effort remains early-stage and may never be commercially viable, while company leaders have argued on-orbit computing could be cheaper to operate than terrestrial facilities. Google’s chief executive has suggested orbital data centers are roughly a decade away from mainstream use.
The talks highlight how quickly on-orbit computing could attract investor, regulatory, and engineering focus if launch demand accelerates. This would intensify scrutiny on launch cadence, orbital traffic management, and the narrow economics that must improve for data centers to move off Earth.





