Google DeepMind·2026-06-16 23:46· 2026-06-16

Securing the future of AI agents

AI summary

Google DeepMind released AI Control Roadmap, a framework for building and managing advanced AI deployed inside Google. It applies defence in depth, adding system-level safety layers beyond model alignment to provide protection even when alignment is imperfect.

Selection record

Threshold 60Official, first-handFirst 62Second 62

AdmittedSum of both 124 ≥ twice the threshold 120

Source tier: Official, first-hand; this tier's threshold is 60
Pre-filter: passed：Google DeepMind AI智能体安全控制框架
Why it was chosen: The internal AI control roadmap details tiered detection and response and findings from analysing a million coding agent trajectories.

A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.

Source: Google DeepMind · deepmind.google

#Research #Safety and alignment #Agents #Google