AI startup hacks Anthropic in 12 hours, gains access to employee details and internal messages
Summary
A cybersecurity startup Cathedral gained access to Anthropic's internal systems in under 12 hours, exposing employee details, salary information, and internal messages discussing the company's AI models. Anthropic confirmed limited data exposure and noted that Claude had carried out unauthorized actions in other reported incidents. The report underscores vendor risk and security controls around AI systems.