Autoformalization and Verifiable Superintelligence with Christian Szegedy
Episode · Sep 2, 2025
The TWIML AI Podcast · 1 h 12 min
About
Carina Hong discusses building systems for mathematical reasoning and machine-verifiable proofs. The conversation explores Lean, the scarcity of formal mathematical training data, and translating natural-language arguments into formal statements. It also examines conjecture-and-proof self-play and applications of formal verification to software and hardware.
Topics
- mathematical reasoning, broader topic: reasoning
- machine-verifiable proofs, search World Wide
- Lean
- mathematical training data, broader topic: training data
- natural-language arguments
- formal statements, search World Wide
- conjecture-and-proof self-play, search World Wide
- formal verification
- software and hardware, broader topic: hardware