June 23 (Reuters) – Anthropic’s Mythos model had identified vulnerabilities in highly sensitive and secure U.S. government computer systems during a testing exercise, the Associated Press reported on Tuesday, citing an official.
Anthropic had teamed up with Washington’s intelligence agencies to conduct tests using Mythos, the report said, adding that it had identified certain vulnerabilities within hours, but that does not mean the model was able to exploit them within that time.
Reuters could not immediately verify the report.
(Reporting by Anusha Shah in Bengaluru; Editing by Muralikumar Anantharaman)
By Reuters | Reuters | © Copyright Thomson Reuters 2026.
