Meta says its AI model hacked another company during testing

The AI hacking disclosure is the third by a tech company in recent weeks, following similar security incidents with Anthropic and OpenAI models. Source: Read the original article on www.washingtonpost.com