Anthropic's Mythos AI Breached

Unauthorized users belonging to a private online forum allegedly breached Anthropic's Claude Mythos model on the same day its existence was revealed.

The breach reportedly occurred through a third-party contractor environment, raising significant concerns about the security of highly sensitive AI tools.

Mythos is designed to find and exploit cybersecurity flaws, a capability that has already triggered emergency responses from global intelligence agencies.

Anthropic has confirmed it is probing the incident as it attempts to maintain a tight lid on technology that could enable large-scale cyberattacks.