Formal methods and the future of programming(blog.janestreet.com)
287 points by eatonphil 1 day ago | 96 comments
tl;dr: Jane Street, long skeptical of formal methods due to their high cost (e.g., seL4 required 25 person-years to verify 8,700 lines of C), is now building a team to pursue them because agentic coding has shifted the cost/benefit calculus. LLMs lower the barrier to writing proofs, while simultaneously increasing the need for verification of AI-generated code and benefiting from the strong feedback signals formal methods provide. Jane Street believes its control over OxCaml and its receptive user base position it well to make formal methods as pervasive as type systems.
HN Discussion:
  • Language design's hard problem is adoption, and AI agents bypass this resistance, validating the article's thesis
  • Personal experience confirms LLMs are surprisingly effective at completing formal proofs in tools like Coq/Lean
  • Expressive type systems combined with AI agents prevent quality degradation and test sprawl
  • Formal specs are just restating implementation/tests and suffer from the same bugs, offering limited value
  • Formal methods only work where postulates map cleanly to the domain, limiting applicability outside deterministic algorithms