As a first step toward building data centers in space, Google partnered with SpaceX today to launch AI chips into space to gauge the potential for more complex ventures. The goal of this mission is to test how Google Tensor Processing Units (TPUs) perform in space.
On a blog post, Google said: Project Suncather, as the mission is called, is “a long-term, research moonshot exploring whether space could one day host scalable machine learning infrastructure. In low Earth orbit, satellites can access near-constant sunlight, generating up to eight times more solar power than on Earth. Eventually, it could be possible to link together multiple constellations of satellites, allowing them to manage larger AI workloads while in orbit.”
The chips will run a version of the company’s Gemma AI model for 15 minutes at a time due to heat management constraints, using the open weight model to answer simple queries. Open weight generally means accessible to anybody, like Google’s Gemini.
This was not the first such AI venture into space. This past November, a company called Starcloud launched a spacecraft that carried an Nvidia H100 chip on board and demonstrated a version of Google’s Gemini AI from space.
The launch comes only two days after President Trump assembled AI business leaders in the White House to sign on to a voluntary safety protocol agreement for future AI advancements.
