Anthropic AI Model Uncovers Vulnerabilities In Classified U.S. Govt Systems During Testing: Report

One of Anthropic's most advanced artificial intelligence models identified vulnerabilities inside highly sensitive U.S.