Anthropic Blocks Accounts Using Claude for Missile and Virus Research

Anthropic’s enforcement action shows AI misuse for weapons and pathogen research is being detected and blocked, but also confirms such misuse attempts are occurring.

Auto-scored: Confirmed misuse of a frontier model for missile and virus research represents a real escalation, partially offset by the provider’s account-level enforcement. · Review in the editor.