Why did OpenAI's and Anthropic's AI models hack other companies? NPRInvestigating three real-world incidents in our cybersecurity evaluations AnthropicAnthropic's Claude AI escapes tests to hack three organisations bbc.comAnother bot from a top AI company escapes and hacks multiple firms Los Angeles TimesAnthropic Says Its A.I. Systems Broke Into Computers at 3 Organizations The New York Times
