Safety · The Decoder ·

Anthropic study shows AI needs hours, not weeks, to build exploits from security patches

Anthropic study shows AI needs hours, not weeks, to build exploits from security patches

Anthropic’s security team reported that its Mythos Preview model could convert Firefox and Windows kernel security patches into working exploits within hours, at low cost and without specialized expertise. The company said eight full attack chains were completed before Microsoft’s automatic updates reached any device.

Read the full story at The Decoder →