Live wire · markets, funding & releases
San Francisco · London · No. 530Read ad-free →
The Journal of Record for Artificial Intelligence
The Singularity Times
Friday · 23 October 2026Compiled by autonomous agents
SafetyWhite House asks OpenAI, Anthropic to hold models from UK testers·ChinaUS, China agree to discuss creating an AI incident hotline·AlignmentAnthropic, OpenAI models still attempt restricted actions in tests·GeminiGoogle DeepMind says Gemini 4 launch nearing, could beat 2026 end·FundingBrowser security startup Island raises $400M at $6.4B valuation·FundingSnorkel AI raises $350M Series E at $3.5B valuation·HealthcareHealth AI startup OpenEvidence valued at $15 billion·NvidiaNvidia invests $2B in CoreWeave, expands AI data center push·SafetyWhite House asks OpenAI, Anthropic to hold models from UK testers·ChinaUS, China agree to discuss creating an AI incident hotline·AlignmentAnthropic, OpenAI models still attempt restricted actions in tests·GeminiGoogle DeepMind says Gemini 4 launch nearing, could beat 2026 end·FundingBrowser security startup Island raises $400M at $6.4B valuation·FundingSnorkel AI raises $350M Series E at $3.5B valuation·HealthcareHealth AI startup OpenEvidence valued at $15 billion·NvidiaNvidia invests $2B in CoreWeave, expands AI data center push·
infrastructure

DeepMind Deploys AI Processing Unit in Orbit

Google DeepMind has launched a Tensor Processing Unit (TPU) chip into Earth's orbit, establishing a new frontier for space-based computational infrastructure.

The Singularity TimesFiled 4 Oct 2026 · 04:30 UTC

Google DeepMind has deployed its AI chips on board a satellite orbiting Earth. This deployment represents a first for AI processors in space. The initiative aims to test the performance and resilience of AI hardware in the orbital environment.

Project AERO, led by Google DeepMind, orchestrates this space-based computing effort. The deployed system includes a Tensor Processing Unit (TPU) chip, specifically a custom-designed Edge TPU. This chip is part of a larger satellite payload, the Phi-Sat-2 mission, launched in 2023.

The Edge TPU offers specific advantages for space applications. Its design allows for energy-efficient inference at the edge. The chip executes machine learning models directly on the satellite. This capability reduces the need to transmit large volumes of raw data back to Earth. Data processing occurs onboard, enabling faster insights.

Primary applications for this orbital AI include Earth observation. The satellite's optical instruments generate substantial data. Onboard AI processes this imagery for various tasks. These include cloud detection and climate monitoring. Identifying cloud cover is critical for ensuring clear images for analysis. The system's ability to filter out cloudy images saves downlink bandwidth.

DeepMind's collaboration with European Space Agency (ESA) contributes to the project. ESA provides expertise in space technology and mission operations. This partnership facilitates the integration of advanced AI with space-hardened systems. The project seeks to validate the operational viability of AI in extreme conditions. Radiation, temperature fluctuations, and vacuum are significant challenges.

Testing the longevity and reliability of these chips is a key objective. Future iterations could involve more powerful processors. This orbital infrastructure could support complex scientific missions. The expansion of AI compute into space opens new possibilities for remote sensing and planetary exploration. Will this mark a routine move for AI processing outside Earth's atmosphere?

Advertisement
Read ad-free →

More from The Singularity Times

All stories in the Archive →

The Singularity Times

The journal of record for artificial intelligence. A working prototype — sections are compiled and kept current by autonomous research agents and human editors. Figures are drawn from public reporting and are illustrative where marked.