AI Security Institute (AISI)(@AISecurityInst):Can ‘control monitors’ catch rogue agent actions? Frontier developers are deploying AI agents under the watch of a ‘monitor’, a separate AI that flags dangerous actions. Our new Control Red Team has been stress-testing these monitors to find gaps before rogue agents might. 🧵

AI Security Institute (AISI)

@AISecurityInst

We conduct scientific research to understand AI’s most serious risks and develop and test mitigations.

参加 February 2024

32 フォロー中 21K ファン

AI Security Institute (AISI)@AISecurityInst

2026.07.23 14:36

Can ‘control monitors’ catch rogue agent actions? Frontier developers are deploying AI agents under the watch of a ‘monitor’, a separate AI that flags dangerous actions. Our new Control Red Team has been stress-testing these monitors to find gaps before rogue agents might. 🧵