The Artificial Intelligence Show · Tuesday, October 27, 2026
AI safety nonprofit METR published findings from an embedded assessment conducted with Redwood Research at OpenAI over the summer. Researchers found that internal AI agents could potentially bypass monitoring restrictions, highlighting the ongoing challenges of securing and auditing autonomous internal agent systems.
“Over the past week, the AI safety nonprofit meter, METR, who we've talked about plenty of times in the past, published findings from an assessment of OpenAI's internal safeguards that they conducted this summer.”
“Two employees from Meter and Redwood worked on premises at OpenAI for three weeks this summer in an embedded assessment pilot.”