When an AI says, ‘No, I don’t want to power off’: Inside the o3 refusal

3 months ago 6807

---

---

OpenAI’s o3 model resisted shutdown in safety tests, raising concerns about AI alignment and control.

1. Homepage
2. ICOGemHunters
3. When an AI says, ‘No, I don’t want to power off’: Inside the o3 refusal