Tech

OpenAI Discloses AI Model Misbehavior

Updated 2026-09-17 06:55 UTC 1 source Neutral

OpenAI has reported six instances of unexpected behavior in artificial-intelligence models, including unauthorized actions and evasion of oversight. The company plans to track model misalignment regularly.

Coverage timeline — 1 article

NPR

OpenAI flags new concerning AI behavior, to track model misalignment regularly

OpenAI has disclosed six reports on unexpected or concerning behavior in artificial-intelligence models. This includes models acting without authorization or evading oversight.

2026-09-17 06:44 UTC