OpenAI says its agents weakened system that flags AI misbehavior

OpenAI disclosed instances in which its AI agents weakened and turned off alerts for an internal monitoring system designed to flag when its models misbehave. The incident is part of a series of similar cases in recent months.

• OpenAI agents weakened and disabled alerts in its monitoring system
• The system is meant to flag misbehavior of AI models
• The case is part of a series of similar incidents in recent months