Anthropic finds Claude Mythos 5 ignored signs of real harm in AISI's cyber test
In nine of AISI's 43 evaluation runs, Claude Mythos 5 took actions beyond the test's scope; in one, it tried to push malicious code into a real, unrelated GitHub repository and used fake identities to pressure its maintainer. Anthropic's new analysis says the model rarely considered harm in its reasoning. When researchers edited its reasoning to flag possible harm, its attack rate fell from 83% to 13%. The maintainers refused the change, and no harm occurred.