Table of Contents
Google is preparing to launch a prototype satellite next week as part of Project Suncatcher, an experimental effort exploring whether large-scale artificial intelligence computing could eventually be carried out in space.
The mission will mark Google’s first in-orbit test of the technology and is expected to provide data on how AI hardware performs under the harsh conditions of low Earth orbit (LEO).
AI Chips to Be Tested in Orbit
The prototype satellite will carry Google’s Tensor Processing Units (TPUs) into low Earth orbit.
The mission is designed to test whether the AI processors can operate reliably while exposed to launch forces, radiation and extreme temperatures.
Google has already tested its TPUs under simulated space conditions while running AI workloads at a facility at the University of California, Davis. However, the company said actual orbital testing is necessary to understand how the hardware performs in space.
Space Radiation Poses Challenges
One of the major challenges is radiation exposure.
Space radiation can interfere with electronic systems and potentially cause data errors known as bit flips. The upcoming mission will help Google identify potential hardware and reliability issues that could affect future space-based computing systems.
Cooling AI Hardware in Space
Google will also test its approach to cooling power-intensive AI chips in the vacuum of space.
Unlike conventional data centers, spacecraft cannot rely on normal airflow-based cooling. The proposed system combines heat pipes and radiators to transfer and release heat generated by the computing hardware.
SpaceX Launch Planned
The satellite is scheduled to fly on SpaceX’s Transporter-18 rideshare mission in partnership with satellite company Planet Labs.
The initial mission is intended primarily to collect data and identify potential technical problems rather than demonstrate a fully operational space-based data center.
Companies Exploring Orbital Data Centers
Google is not alone in investigating the possibility of moving AI computing infrastructure into space.
Companies including SpaceX and Starcloud are pursuing concepts involving data centers in low Earth orbit. Proponents of the technology point to the availability of near-continuous sunlight as a potential source of power for energy-intensive computing.
However, experts have highlighted significant challenges, including launch costs, engineering requirements and satellite manufacturing capacity.
Google Plans Further Testing
Google plans to conduct additional tests as part of Project Suncatcher.
The company aims to launch two satellites in 2027 to test high-bandwidth laser communication links that could eventually connect computing clusters in orbit.
The upcoming mission will therefore serve as an early step in evaluating whether space-based AI computing can become technically and commercially practical.
