Rich Sutton and Khurram Javed: Why AI Models Stop Learning, and How to Start It Again
Rich Sutton and Khurram Javed founded Oak Lab to solve AI's continual learning problem
“All learning is continual. We always act and we learn. That's just the normal way of thinking.”
Reinforcement learning pioneer Rich Sutton and his former student Khurram Javed have co-founded Oak Lab, targeting the fundamental failure of modern AI models to keep learning after training. Sutton argues the field has lost its way — 'continual learning' is redundant because all real learning is continuous, and the current batch-training paradigm is the anomaly. Coming from the author of 'The Bitter Lesson,' this is a direct foundational critique of the LLM-centric AI moment and a signal that serious researchers are building alternatives.