← Back to feed

A critical look at Lean4

L5 · ResearcherOpinion & AnalysisLessWrong AI· 10/9/2026

Addresses foundational concerns about formal verification tools used in AI safety research and certification.

AI Summary

Critically examines Lean4's limitations as an immature formal verification tool, questioning its readiness for societal-level trust in AI safety certification.

Excerpt

Lots of the discussion around Lean certificates and trust centers around general philosophical ideas of how much humans should trust formal certificates. It's important to remember, though, that Lean4 is not an ideal formal certificate machine: Lean4 is a software in its infancy and has issues. There are reasonable reasons to specifically not trust Lean4. If we're going to be putting our society's trust in Lean4, I think that it's good to have an open critical discussion of how warranted that tr

Read Original
0 upvotes · 0 downvotes · 1 min read

Related Articles

L3 · BuilderOpinion & AnalysisHacker News
Why isn't the industry freaking out about DeepSeek 4.1 Flash?

DeepSeek 4.1 Flash performs comparably to frontier models at dramatically lower cost, making high-volume AI development affordable for builders.

L5 · ResearcherOpinion & AnalysisHacker News
OpenAI, the Partition Principle, and Mathematics

A mathematician critiques OpenAI's preprint claiming to prove the Partition Principle doesn't imply the Axiom of Choice, calling it poorly written with strange structure and questionable references.

L4 · DeveloperOpinion & AnalysisLessWrong AI
Stop telling people to pivot to AI safety, I beg you

An AI safety researcher argues against calls for more people to pivot into AI safety, citing a severe job shortage where talent supply significantly exceeds available positions.

L2 · PractitionerOpinion & AnalysisMIT Technology Review AI
AI breakthroughs in robotics won’t change your life any time soon

MIT Technology Review argues that AI-powered humanoid robots like Tesla's Optimus face significant technical challenges and won't deliver on near-term promises.

L5 · ResearcherOpinion & AnalysisLessWrong AI
ASI and Liberty: Can we do better than a benevolent god?

Zvi distinguishes between AGI-pilled and ASI-pilled perspectives on AI progress, questioning whether superintelligence could rapidly advance technology to cosmic scales.