Anthropic AI risks warned

A new AI model from startup Anthropic, dubbed Claude Mythos, has triggered an emergency response from international financial watchdogs.

The model's advanced capability to identify zero-day vulnerabilities has led the U.S. government to issue a directive to Wall Street banks to test the system for defensive purposes.

In the U.K., the Financial Conduct Authority and the National Cyber Security Centre are briefing major insurers and exchanges on the potential for the AI to upend digital security.

Critics and industry insiders are debating whether the model represents a genuine breakthrough in autonomous hacking or a calculated pre-IPO hype campaign.