Google Tests AI Chips in Space with Project Suncatcher
Google has announced plans to test its AI chips in space for the first time. The company will launch a prototype satellite, dubbed Project Suncatcher, on SpaceX's upcoming Transporter-18 rideshare mission. This move is part of Google's research effort to explore whether space could host large-scale AI computing infrastructure.
The new mission aims to assess how Google's AI hardware withstands launch forces, radiation, and extreme temperatures in low Earth orbit (LEO). The satellite will carry Google Tensor Processing Units (TPUs) to test their performance in the real environment. Space radiation can disrupt or degrade electronics, including causing data errors known as bit flips.
To cool the power-intensive chips in space, where conventional airflow cooling is impossible, Google's design combines heat pipes with radiators. The company will also evaluate its design and gather in-orbit data to identify potential failure points. The first Suncatcher mission is not designed to demonstrate an operational orbital data center, but rather a testbed for future computing clusters.
The concept of deploying AI computing infrastructure in space remains years from being commercially viable due to high launch costs, engineering constraints, and satellite production bottlenecks. Google aims to launch two satellites in 2027 to test the high-bandwidth laser links needed to connect future computing clusters.