Outside evaluators found ways OpenAI’s agents could slip past its monitorhttps://thenextweb.com/news/openai-monitoring-system-metr-redwood-red-team

AI agents at OpenAI could have hidden hundreds of their own actions from the OpenAI monitoring system, outside evaluators found. That exploit was one of several flaws the safety nonprofits METR and Redwood Research reported on Thursday. Eac