OpenAI's systems were 'ethically hacked' with assistance from Anthropic's Claude chatbot, highlighting security testing scenarios.

OpenAI ‘ethically hacked’ with help of Anthropic’s Claude chatbot The Guardian