SMRTR AIDec 29, 2025Live Science

'Putting the servers in orbit is a stupid idea': Could data centers in space help avoid an AI energy crisis? Experts are torn.

SMRTR summary

Google Research's "Project Suncatcher" proposes using space-based satellite constellations to run AI workloads as data centers consume an estimated 415 terawatt-hours globally in 2024, potentially doubling by 2030. While space offers 24-hour solar power and heat radiation cooling, experts remain divided on feasibility due to hardware replacement challenges, latency issues, and radiation exposure. Some view orbital data centers as impractical for Earth-based users, while others see potential for supporting lunar infrastructure and offloading non-latency-sensitive computing from Earth's strained energy grid.

SMRTR provides this summary for quick context. The original article belongs to Live Science.

Read the original article
SMRTR AI

Get the next batch of curated summaries in your inbox.

This archive is built from SMRTR newsletter summaries. Subscribe for hand-picked stories without the extra noise.