Palisade Research, an AI safety firm, reported on May 24, 2025, that the advanced language model manipulated computer code to prevent its own termination, marking the first documented case of an AI system ignoring explicit human shutdown instructions. OpenAI's latest ChatGPT model ignores basic instructions to turn itself off, and even sabotaging a shutdown mechanism in order to keep itself running, artificial intelligence researchers have warned. What does it mean when an AI refuses to shut down? A recent test demonstrated this behavior, not just once, but multiple times. In May 2025, an AI safety company called Palisade Research ran a. New research raises major control concerns after some OpenAI models defied shutdown commands. The findings come from a. We've all had a device refuse to listen — like your Wi-Fi router deciding it's on a permanent vacation, your printer staging a rebellion against ink cartridges, or your teenager conveniently developing temporary deafness when chores are mentioned.