Anthropic’s Claude Aided Researchers in OpenAI Breach

Three researchers reportedly used Anthropic’s Claude AI model to gain unauthorized access to OpenAI’s internal systems within 72 hours.