threat-intel
Anthropic’s restricted Claude Mythos model may be coming to Claude Code
High
Summary
Anthropic is preparing to release a new AI model called Mythos, initially designed for advanced computer security tasks. The model demonstrates a concerning ability to autonomously develop cyberattacks, raising significant security risks. Anthropic is proactively addressing these risks through a collaborative project, Glasswing, and a robust guardrail system, but the potential for misuse remains a key concern.
Summary written automatically in our own words from the original article, which belongs to its publisher and remains the reference. It may contain errors. Sources & data