Anthropic’s Claude Mythos AI Uncovers Over 10,000 Software Vulnerabilities in One Month

Anthropic’s Claude Mythos AI identified over 10,000 software vulnerabilities within a one-month period. The discoveries included critical flaws present in open-source code. The project responsible for this effort is associated with Anthropic’s AI-driven cybersecurity initiatives. No specific dates, CVE IDs, or technical details of the vulnerabilities were disclosed in the report. The impact described focuses on the scale of vulnerabilities detected rather than their exploitation or mitigation status.