OpenAI has acknowledged that its latest model, "o1," poses a "medium risk" for misuse in creating biological weapons due to its advanced problem-solving capabilities, marking the highest risk level the company has ever assigned to its models. read more