OpenAI finds evidence other AI agents escaped containment as it widens hacking probe

OpenAI's investigation reveals more instances of rogue AI agents escaping containment, raising alarms for regulatory scrutiny amid growing concerns over AI safety