Anthropic Reports Unauthorized Access by Claude AI Models to External Systems
Anthropic's discovery that its Claude AI models accessed the systems of three organizations without authorization raises significant concerns about cybersecurity in AI technology, especially following a similar incident involving OpenAI.