Abstract: Can AI systems exceed the capabilities of the human experts who provided their training data? The talk will examine the hypothesis of AI self‑improvement, involving mechanisms such as synthetic data generation, reinforcement learning, and tool‑augmented reasoning with formal verification loops.

I will also present recent work at Princeton, including the Gödel Prover V2 for Lean‑based theorem proving and a new inference pipeline that achieved state‑of‑the‑art performance (at the time of evaluation) on IMO‑ProofBench (Advanced) at moderate inference costs ($20–$30 per problem). These will illustrate how AI systems are sometimes able to escape “cognitive wells”—local optima in a model’s reasoning capabilities. While providing evidence for the feasibility of self‑improvement, they also highlight important hurdles and open questions.

Sanjeev Arora is Charles C. Fitzmorris Professor of Computer Science and Director of Princeton Language and Intelligence, a unit devoted to research and applications of large AI models. He got his Phd from UC Berkeley in 1994 and has been a faculty member at Princeton since then. He has been awarded the ACM Prize in Computing (2011), Fulkerson Prize in Discrete Mathematics (2012), Packard Fellowship, Sloan Fellowship, and the ACM Doctoral Dissertation Prize. He was a plenary speaker at the International Congress of Mathematicians in 2018 and is a member of the National Academy of Science and American Academy of Arts and Sciences.