Google’s proprietary artificial intelligence chips reached orbit on Thursday following a launch aboard a SpaceX Falcon 9 rocket, signaling a major step toward establishing data centers in space. The mission, designated Transporter-18, lifted off from Vandenberg Air Force Base in California at 11:15 a.m. PT and successfully deployed all payloads by 1 p.m. local time.
The flight carries Planet Labs satellites, including a solar-powered prototype equipped with Google’s tensor processing units (TPUs). This deployment serves as the first in-orbit test for Project Suncatcher, an initiative Alphabet announced in November 2025. The project aims to create reliable, solar-powered AI computing infrastructure capable of continuous operation in the harsh environment of low Earth orbit.
Travis Beals, senior director of Project Suncatcher, confirmed that his team has established contact with the satellite and it is functioning as expected. In a blog post, Beals described the launch as the initial phase of a long-term research endeavor to determine if scalable machine learning infrastructure can eventually operate in space.
Alphabet highlighted that satellites in low Earth orbit can access near-constant sunlight, potentially generating up to eight times more solar power than systems on the ground. The company plans to link multiple satellite constellations in the future to handle larger AI workloads while in orbit.
The collaboration underscores the close ties between the two tech giants, despite some competitive overlap in the AI sector. Alphabet holds a stake in SpaceX valued at more than $82 billion following the rocket manufacturer’s record-breaking public offering in June. SpaceX, led by Elon Musk, has also indicated plans to develop its own orbital data centers using swarms of satellites equipped with graphics processing units.
While Google has previously tested its TPUs running AI workloads at the University of California, Davis, the company is now evaluating how the chips perform under the specific conditions of low Earth orbit.
Can someone explain why TPU performance differs in low Earth orbit compared to a grounded server farm?
Does this mean cheaper compute costs for everyone, or just another way for Big Tech to monopolize resources?
I never thought I’d see Google and SpaceX collaborating so closely on infrastructure. The space race is heating up!
Feasible in theory, but what happens when that hardware fails three years into its life up there?
Eight times more solar power? That changes the economics of running massive AI models in orbit completely.