Who Checks a Proof No Human Can Read? — Leo de Moura
Machine Learning Street Talk (MLST)
Leo de Moura, creator of the Lean programming language, explores the intersection of formal verification and artificial intelligence. Lean, a tool for verifying mathematics and software, relies on a "cathedral" development model to maintain core integrity while allowing community-driven extensions. AI significantly accelerates the formalization process, enabling the translation of complex code and the discovery of bugs in kernels, though it requires human oversight to ensure structural soundness. The conversation highlights the shift toward specification-driven engineering, where AI handles the labor-intensive task of updating proofs and optimizing code. As Lean scales, the focus remains on reducing the trusted codebase and leveraging AI to manage increasingly large mathematical libraries, ultimately transforming how developers approach software reliability and mathematical discovery.
Sign in to continue reading, translating and more.
Open full episode in Podwise
