Anthropic says its AI systems blocked users trying to obtain bioweapons knowledge

Anthropic says it blocked possible efforts to build biological weapons

Anthropic published a threat intelligence report claiming its safety systems detected and blocked users attempting to use Claude for bioweapons knowledge. The post doesn't disclose technical details, number of users involved, or timeline. Only the headline and snippet are available—I'd wait for the full report before assessing how effective the blocking actually was.