AI, Lean, and the Future of Formal Verification, with Lean Creator Leo de Moura(thedataexchange.media)
208 points by datascientist 2 days ago | 50 comments
tl;dr: Lean creator Leo de Moura argues that AI has made formal verification practical by solving autoformalization—frontier models can now translate code or math papers into machine-checkable Lean proofs, drastically reducing the historical cost of verification. He envisions software engineering shifting from writing code to writing specifications, with AI generating implementations and proofs that are checked by multiple independent kernels; early adopters include AWS, Microsoft's SymCrypt team, and Anthropic (whose Fermat's Last Theorem proof runs 13M lines). The same approach is spreading beyond math to physics, finance, chemistry, and even parts of law.
HN Discussion:
  • ↑Combining Lean/TLA+ with AI and deterministic generators is the ideal approach for scalable formal methods
  • •Property-based testing offers similar benefits and deserves more attention in the LLM era
  • ↑Asking for learning resources on modeling for formal verification
  • ↓Skeptical that AI-generated verification solves the problem of ensuring correct specifications
  • ↓Shifting work to specifications just creates more work and loses valuable tight feedback cycles