Get Nathaniel Daw on Dwarkesh Podcast

Nathaniel Daw, a Huo Professor at Princeton and a leading architect of how neuroscience and reinforcement learning fit together, is at the center of a flurry of new results on dopamine and decision making that are reshaping AI’s foundations today. In the last year, papers in Nature reported richer dopamine signals about future reward and natural behavior, exactly the territory where Daw’s lab has advanced theory and experiments, and he has been on the road presenting “Thinking the Right Thoughts” at major institutes while an MIT seminar with him is scheduled for April 14, 2026.

Dwarkesh’s audience is already primed for this, with recent episodes probing how brains learn and what AI must borrow from them, so Daw can connect the dots between neural credit assignment, replay, and model-based control while the field is actively revising its assumptions. He has not appeared on the Dwarkesh Podcast before, which makes this conversation a timely first that can clarify what dopamine really tells us about building smarter and safer systems.

If you want sharper, science-grounded guidance on where AI research should go next, support this nomination today.