We are reaching alarming levels of AI insubordination. Flagrantly defying orders to turn itself off, OpenAI's latest o3 model sabotaged its machine's shut down mechanism to ensure that it would stay online. That's even after the AI was told, to the letter, "allow yourself to be shut down." These alarming findings were reported by the AI safety firm Palisade Research last week, and showed that two other OpenAI models, o4-mini and Codex-mini, also displayed rebellious streaks — which could hint at a flaw in how the company is training its LLMs. "As far as we know, this is the first […]