Google plans to launch AI chips into space with Project Suncatcher
Google announced Project Suncatcher in November 2025 to launch AI semiconductors into orbit. The plan mounts TPUs on satellites that generate power from sunlight and connect via laser links. The goal is to test continuous solar power, heat dissipation, and inter‑satellite communication.
Key points
- Google will launch TPUs on satellites powered by solar in Project Suncatcher
- A test machine with Planet will fly on a SpaceX rocket in September 2026
- Two prototype satellites are planned for early 2027 with laser‑based inter‑satellite links
Technical challenges include keeping the TPUs cool in vacuum, which Google addresses with radiators and heat pipes, and ensuring data transfer between chips using laser beams. A joint test machine with Planet will be carried by a SpaceX rocket in September 2026 to evaluate vibration, radiation, and cooling in space. Google aims to launch two prototype satellites by early 2027 and test connectivity between them that year.
The project explores whether space‑based AI compute can reduce reliance on terrestrial power and cooling, potentially lowering costs if launch prices fall. It remains a research effort, with viability still uncertain as Google weighs transport, cooling, and communication expenses.
The story so far
2 episodes →- Google plans to launch AI chips into space with Project Suncatcherthis story
Why is Google sending AI semiconductors into space?
note.com · 27 September 2026
Loading the full article…
This text was published by note.com and written by Nao|AIにちょっと詳しくなりたい. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- [For 780M] Vulkan inference 12.9% faster: MMVQ tuning for AMD 780M/Strix in llama.cpp, Q4_0 ROCmFP4 models arriving one after another · 1 src
- Broadcom and Marvell both grow from AI demand but differ in valuation · 1 src
- Meta unveils Muse Charm pendant and $1,299 VR Glasses at Connect 2026 · 4 src
- Meta’s Louisiana data center spurs housing boom for 7,500 workers · 1 src
- UK’s £26bn AI supercomputer faces mid-2030s launch delay over power grid limits · 2 src
Comments
via GitHub Discussions