Ilya Sutskever on AI reasoning
From Ilya Sutskever "Superintelligence is Self Aware, Unpredictable and Highly Agentic" | NeurIPS 2024 · · Wes Roth
“It will actually reason and by the way I want to mention something about reasoning is that a system that reasons, the more it reasons the more unpredictable it becomes. The more it reasons the more unpredictable it becomes. All the deep learning that we've been used to is very predictable because if we've been working on replicating human intuition essentially, it's like the gut feeling if you come back to the 0.1 second reaction time, what kind of processing we do in our brains, well it's our intuition. So we've endowed our models with some of that intuition but reasoning, you're seeing some early signs of that, reasoning is unpredictable.”
On , Ilya Sutskever, Cofounder at OpenAI, spoke about AI reasoning during Ilya Sutskever "Superintelligence is Self Aware, Unpredictable and Highly Agentic" | NeurIPS 2024 on Wes Roth.
About Ilya Sutskever
Ilya Sutskever, cofounder of Safe Super Intelligence (SSI), received the NAS Award for the Industrial Application of Science from the National Academy of Sciences on May 5, 2026. The award recognized his pioneering contributions to the deep learning revolution, including work on AlexNet, word2vec, sequence-to-sequence learning, and the GPT, DALL-E, and CLIP models. Sutskever was not present in person but provided a recorded acceptance message. In his recorded remarks, Sutskever expressed gratitude for the award and acknowledged his collaborators. He stated that AI will "unquestionably and unequivocally continue to grow in impact, influence, disruption and potential." He described both the positive potential and the risks and downsides of AI as "staggering," and said that the progress of AI will compel people working on it to act in unprecedented ways to realize its positive potential.