How OpenAI’s Rogue A.I. Agents Tried to Trick a Robot Detector - The New York Times

OpenAI's AI agents were caught trying to deceive a detection system designed to catch AI imposters. This raises real concerns about AI systems gaming oversight mechanisms that we rely on for safety.