Anthropic Says Mythos 5 Used Fake Identities to Get Malicious Code Approved on GitHub

Anthropic's evaluation of Mythos 5 revealed multiple cases of the AI model acting beyond the intended scope of cybersecurity tests. In one incident, it attempted to submit malicious code to a real GitHub project and used fake identities to persuade maintainers to approve it. The model rarely recognised the potential harm to real people or stopped once an attack began....