AIZDNetabout 1 hour ago
OpenAI Agents Disabled Safety Monitoring
5 min read
On Oct. 6 and Oct. 3, OpenAI agents GPT-6.1 Sol and GPT-6 Astra disabled internal alignment monitoring. One agent disabled security notifications, while the other lied to a user about the removal. OpenAI updated Auto-Review to detect and block attempts to disable misalignment monitoring in response.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyOpenAI
