Disrupting malicious uses of AI: June 2025

In three linesOpenAI releases report on detecting and preventing malicious AI uses with case studies. No technical details or metrics provided in excerpt.

Summary generated by Claude — human-verified