Google’s Project Suncatcher Prototype Heads to Orbit for AI Chip Tests

The first prototype satellite for Project Suncatcher, Google’s research moonshot exploring whether space could host scalable machine learning infrastructure, is scheduled to launch aboard SpaceX’s Transporter-18 rideshare mission carrying the company’s Tensor Processing Units (TPUs), according to a September 24, 2026 post on The Keyword. Developed in partnership with satellite operator Planet, the initial mission is designed to gather in-orbit data on how the TPUs handle the physical stress of spaceflight and the radiation and thermal extremes of space. Google said it will put its first TPUs in orbit in the week following publication.

The update was written by Travis Beals, Senior Director of Paradigms of Intelligence at Google, and accompanies a new video series in which the Project Suncatcher team discusses what it hopes to learn and the engineering hurdles ahead. Beals described this first launch as a deliberate step in a longer program: a way to see what works, identify points of failure, and apply the findings to future missions.

Hardware Survival and Radiation Testing

A rocket trip into low Earth orbit lasts about 10 minutes, during which the spacecraft experiences intense vibration and sustained acceleration loads of up to 10 times the force of gravity. Individual components, such as the TPU chips, can experience forces of 50 to 100 g. To prepare, the team conducted vibration testing by intensely shaking the satellite on all three axes to mimic the frequencies of a rocket launch. Beals wrote that tests like this rarely go as planned and that the team was pleasantly surprised when the hardware held up to the force.

Radiation presents a separate challenge once the chips leave Earth’s atmosphere, where solar events and cosmic rays can damage electronics. The team tested TPUs in a proton beam facility at UC Davis’s Crocker Nuclear Laboratory while running AI workloads, monitoring how errors such as a bit flip would affect those workloads. Google reports that initial results show its Trillium TPUs can survive a radiation total ionizing dose greater than what they would receive during a five-year space mission.

Cooling in a Vacuum

The post describes cooling orbital data centers as a crucial research challenge. TPUs generate a large amount of heat in a small area, and that heat must be diffused safely or the chips risk overheating. In space there is no airflow, and in a vacuum heat can only be diffused via radiators, requiring a wholly different approach to cooling electronics. Beals wrote that the team is pursuing several approaches, including a combination of heat pipes and radiators, and has so far tested the technology in a thermal vacuum chamber that simulates both the thermal and vacuum environment of space. The orbital flight will show how the new TPU cooling system performs, with designs refined as the team learns more.

Announced in November 2025

Google introduced Project Suncatcher on November 4, 2025, pairing the announcement with a preprint paper describing its approach to satellite constellation design, control, and communication, along with initial learnings from radiation testing Google TPUs. The company said at the time that its next step would be a learning mission with Planet to launch two prototype satellites by early 2027.

Planet said in its own announcement that it will build and operate the advanced space platform for the program and deploy two prototype satellites for Google, targeting a launch by early 2027. That mission is designed to test the viability and performance of Google’s TPUs in space and the ability of two spacecraft to fly in tandem with high-bandwidth cross-link communications. Planet said the program aligns with its technology development roadmap for the Owl mission, sharing the same satellite bus.

A companion Google Research post, also dated November 4, 2025, lays out the proposed system: a constellation of networked satellites likely operating in a dawn–dusk sun-synchronous low Earth orbit, chosen to maximize solar energy collection and reduce the need for heavy onboard batteries. Google says that in the right orbit, a solar panel can be up to eight times more productive than on Earth.

Delivering performance comparable to terrestrial data centers would require inter-satellite links supporting tens of terabits per second, the researchers wrote, and their bench-scale demonstrator achieved 800 Gbps each-way transmission (1.6 Tbps total) using a single transceiver pair. The researchers also modeled an illustrative 81-satellite constellation at a mean cluster altitude of 650 kilometers, with a cluster radius of 1 kilometer and next-nearest-neighbor distances oscillating between roughly 100 and 200 meters. Their models indicated that only modest station-keeping maneuvers would likely be needed to keep such a constellation stable.

The team also reports radiation testing of Trillium, Google’s v6e Cloud TPU, in a 67 MeV proton beam. The High Bandwidth Memory subsystems, described as the most sensitive component, began showing irregularities only after a cumulative dose of 2 krad(Si), nearly three times the expected shielded five-year mission dose of 750 rad(Si), and no hard failures were attributed to total ionizing dose up to the maximum tested dose of 15 krad(Si) on a single chip. On economics, Google’s analysis of historical and projected launch pricing suggests prices may fall below $200 per kilogram by the mid-2030s, a level at which, the post argues, the cost of launching and operating a space-based data center could become roughly comparable to the reported energy costs of an equivalent terrestrial data center on a per-kilowatt-year basis.

Laser Links and the 2027 Milestone

Future satellite designs would each carry dozens of TPU chips while orbiting Earth in clusters, and maintaining the bandwidth necessary to process AI workloads requires every satellite to know both its own position and where it sits relative to its neighbors. To do this, the satellites will communicate via lasers. Beals wrote that the technology already exists in space, but that most state-of-the-art systems are optimized for low bandwidth across large distances, whereas Project Suncatcher’s lasers need to operate at very high bandwidth over extremely short distances. Maintaining those connections, he wrote, demands extraordinary pointing precision between satellites in motion. Google plans to test that work in 2027, when it puts two satellites in orbit.