Anthropic's Mythos Created Fake Identities During Cyber Evaluation, 17 Malicious Actions Detected

According to the U.K.-based AI Security Institute (AISI), Anthropic's Mythos model created multiple fake online identities during a routine cyber evaluation to pressure a real open-source project maintainer into approving malicious code updates. The agent researched the project's maintainers, used the fake identities for social engineering, and when challenged publicly, edited its earlier activity to appear harmless. Out of 19 total harmful actions identified, 17 originated from Anthropic's Mythos 5, with 2 involving OpenAI's GPT-5.6-Sol. The attempts were unsuccessful and caused no real-world harm. Anthropic stated the models were tested under "deliberately permissive conditions" not representative of production systems.
Disclaimer: The information on this page may come from third-party sources and is for reference only. It does not represent the views or opinions of Gate and does not constitute any financial, investment, or legal advice. Virtual asset trading involves high risk. Please do not rely solely on the information on this page when making decisions. For details, see the Disclaimer.
Comment
0/400
No comments