Disrupting malicious uses of AI | February 2026

Source: OpenAI Blog ·

Our latest threat report examines how malicious actors combine AI models with websites and social platforms—and what it means for detection and defense.