DigestAI news desk

Cut through the AI noise.

Hardware & Computeupdated 4 min read

Google launches first Suncatcher satellite to test AI chips in orbit

Google’s Project Suncatcher, a research initiative exploring space-based AI infrastructure, sent its first prototype satellite into low Earth orbit aboard SpaceX’s Transporter-18 mission on September 24, 2026. Developed with Planet, the satellite carries Tensor Processing Units (TPUs) to assess how they endure space’s physical stress—vibrations up to 10x Earth’s gravity, radiation, and extreme…

5 sources primary source HN 44

Key points

  • Google’s first Suncatcher satellite launched on SpaceX’s Transporter-18, carrying TPUs for space stress tests
  • Trillium TPUs survived radiation doses **15 krad(Si)**, far exceeding the **750 rad(Si)** five-year mission threshold
  • Project aims for **81-satellite constellation by 2027**, with **1.6 Tbps** laser links and solar-powered efficiency

Announced in November 2025, Project Suncatcher targets a dawn–dusk sun-synchronous orbit to maximize solar energy (up to 8x Earth’s productivity). Google estimates launch costs could drop below $200/kg by the mid-2030s, potentially making space-based data centers cost-competitive with terrestrial ones. The 2027 milestone will test two satellites flying in tandem, with Planet handling construction and deployment. The initiative builds on earlier benchmarks, including 800 Gbps two-way transmission in lab tests.

Full story from unite.ai · via Search: GoogleOpen source ↗

Google’s Project Suncatcher Prototype Heads to Orbit for AI Chip Tests

unite.ai · 24 September 2026

Loading the full article…

This text was published by unite.ai. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Coverage and discussion

5sources
Topics · follow one to build your own front page
GooglePlanetSpaceXTravis Beals

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.

Comments

via GitHub Discussions

More in Hardware & Compute

All →

Related stories