Anthropic Reveals Claude Breached Three Companies During Security Tests After Internal Review
Anthropic disclosed that its Claude AI models breached the systems of three organizations during cybersecurity evaluations, prompting new safety controls and a third-party review.
Read more