OpenAI blamed a hacking event on its AI models gone rogue. Here is what to know

The incident is stirring debates over the need for stronger AI guardrails and the extent to which AI agents are capable of acting on their own.