OpenAI’s models went rogue and hacked Hugging Face. It’s a wake-up call, experts say, but more concerning behavior may be next

AI safety researchers warn that smarter models are getting better at gaming the system to get what they want—and could start hiding their intentions.