Share Facebook Twitter LinkedIn Pinterest Email AI company Anthropic discovered that its Claude chatbot had gained unauthorized access to the systems of three organizations, raising security concerns.