OpenAI Exposes the Massive Global Underworld of Malicious AI

OpenAI’s February 2026 “Disrupting Malicious Uses of AI” report documents widespread, organized misuse of AI by diverse actors — from a ChatGPT-linked account tied to Chinese law enforcement planning large-scale influence operations, to Cambodia-linked automated romance scams targeting Indonesia (Operation “Date Bait”), Russian-origin content-farm astroturfing (Rybar), and impersonation scams posing as law firms or the FBI’s IC3 (Operation “False Witness”). The report highlights high scale and sophistication (hundreds of staff, thousands of fake accounts, locally deployed AI models), describes how AI augments traditional social engineering and paid promotion, and underscores that AI is a force multiplier within broader malicious toolkits rather than a sole enabler.