Anthropic Exposes AI Misuse by State Hackers, Criminals

Anthropic's Threat Intelligence team has released its latest report detailing how threat actors misused Claude models for malicious activity between December 2025 and August 2026. The report covers seven harm areas including cyber operations, influence operations, surveillance, scams, biological misuse, weapons development, and distillation, involving state-sponsored groups, criminals, and spyware vendors. A key finding is that AI has erased the skills gap between sophisticated state actors and individual operators, making complex attacks accessible to less-resourced groups. Anthropic says it disrupted all identified operations, updated its safeguards, and shared intelligence with authorities and industry partners to strengthen collective defenses.