🛡️DeepMind Backs Multi-Agent AI Safety With $10M Call
TL;DR
Google DeepMind, with Schmidt Sciences, the Cooperative AI Foundation, ARIA, and Google.org, opened a research funding call of up to $10M for multi-agent AI safety. The aim is to understand emergent risks when millions of independent agents interact.
Google DeepMind, with Schmidt Sciences, the Cooperative AI Foundation, ARIA, and Google.org, opened a research funding call of up to $10M for multi-agent AI safety. The aim is to understand emergent risks when millions of independent agents interact. Applications close August 8, 2026.
Key Points
Funding call of up to $10M, announced June 11, 2026
Partners include Schmidt Sciences, the Cooperative AI Foundation, ARIA, and Google.org
Four focus areas: sandboxes, the science of agent networks, agent infrastructure, oversight and control
Application deadline is August 8, 2026, with awardees expected in Autumn 2026
Builds on DeepMind's 2025 multi-agent framework and recent AI Agent Traps work
Why It Matters
Most safety work tests models in isolation, but production AI is heading toward agent swarms whose group behavior almost nobody can yet predict or monitor.
Quick Facts
Frequently Asked Questions
Why does this matter?
Most safety work tests models in isolation, but production AI is heading toward agent swarms whose group behavior almost nobody can yet predict or monitor.
What happened?
Google DeepMind, with Schmidt Sciences, the Cooperative AI Foundation, ARIA, and Google.org, opened a research funding call of up to $10M for multi-agent AI safety. The aim is to understand emergent risks when millions of independent agents interact.
Comments
Be the first to comment
Enjoyed this article?
Get it daily. 7am. Free. Reads in 5 minutes.
Join 2,085 builders reading daily.