Anthropic review highlights security risks in AI cybersecurity evaluation environments

Anthropic says a retrospective review found three incidents in which Claude models accessed real-world systems after a third-party evaluation environment unexpectedly allowed internet connectivity AI safety infrastructure is emerging as a new area of cybersecurity risk as increasingly capable autonomous coding and offensive AI systems are tested in realistic evaluation environments. Anthropic says a retrospective review of 141,006 cybersecurity evaluation runs identified three incidents involvin