Outside testers found flaws in OpenAI's system for watching its AI
Two AI safety groups, METR and Redwood Research, tested OpenAI's internal monitoring system this summer. Each sent one researcher with deep access to the company's code and staff. They found several new flaws. One could have let AI agents hide "hundreds" of actions from monitoring. OpenAI says it has fixed the flaws that affect its safeguards.
Where this came from. 2 outlets reported this. We read them, then wrote the sixty words above.