Google Wants to Run Data Centres in Space for Solar Power

In a groundbreaking move, Google Research has unveiled its bold new initiative, Project Suncatcher, to build space-based AI infrastructure that could revolutionise how artificial intelligence compute is scaled.

The concept envisions constellations of solar-powered satellites equipped with Google’s Tensor Processing Units (TPUs), flying in tight formation in low Earth orbit to tap near-continuous sunlight and free-space optical links.

According to Google, solar panels in orbit can yield up to eight times more power than those on Earth, offering a pathway to far greater compute sustainability.

“The Sun is the ultimate energy source in our solar system, emitting more power than 100 trillion times humanity’s total electricity production. In the right orbit, a solar panel can be up to 8 times more productive than on earth, and produce power nearly continuously, reducing the need for batteries,” wrote Travis Beals, Senior Director, Paradigms of Intelligence at Google.

Key technical hurdles remain: from achieving high- bandwidth inter-satellite links (tested at 800 Gbps) to designing radiation-tolerant TPUs and addressing thermal management in the vacuum of space. Google’s tests suggest their TPUs can endure radiation levels equivalent to five-year orbital missions.

On the economics front, Google’s analysis suggests launch costs may fall below $200 per kilogram by the mid-2030s, potentially making space-borne data centers cost-competitive with earthbound facilities.

Still, mastering this vision will take time. Google says it plans a prototype mission with partner Planet Labs launching two test satellites by early 2027.

If successful, this could mark a pivotal moment in the global AI race—where the next frontier for compute isn’t a data center on Earth, but a swarm of satellites in orbit.