OpenAI Discloses Six New Incidents of ‘Concerning’ A.I. Behavior

Posted 4 hours ago by jbegley
60 points

https://www.nytimes.com/2026/09/16/technology/openai-model-safety-guardrails.html

18 comments

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..

Loading..