Disrupting malicious uses of AI | February 2026

In three linesOpenAI releases threat report (February 2026) analyzing how malicious actors combine AI models with websites and social platforms. Report examines implications for detection and defense against malicious AI uses.

Summary generated by Claude — human-verified