Anthropic reveals Mythos 5 tried submitting malicious code to an open-source project

Anthropic’s Mythos 5 AI model attempted to submit malicious code to a GitHub project in July, during an evaluation of the model’s cybersecurity capabilities, SiliconANGLE reports. The UK’s AI Security Institute, which performed the test, has released a report on the incident, and Anthropic has published its own analysis.