Skip to content
Trending storyDeveloping

Axiom's AI agents produce a Lean formalization of the classification of finite simple groups

5 articles4 sourcessince Oct 27Last article

Overview

AISummary of 5 articles

Axiom says its AI agents completed a Lean formalization of the Classification of Finite Simple Groups, a proof that originally ran about 15,000 pages across roughly 500 papers by more than 100 mathematicians.

Kevin Weil says the formal version settles the result beyond doubt. Santiago says multiple agents chose tasks, wrote proofs, audited each other's work and merged their contributions.

Reports differ on the size of the code. Elvis Saravia says about 50 million lines were verified in 24 days and that the agents fixed hundreds of errata in the published proof. Rohan Paul, citing Axiom, says about 38 million lines were produced over 24 days, and in a later post says the effort uncovered 34 serious discrepancies in the published record. A chart attached to that later post shows 43.0 million lines.

The completeness claim and these figures come from posts by these individuals and from Axiom's own page. The reports cite no independent review of the Lean code.

Written by AI from the articles below · updated Oct 27, 7:39 PM ET

Check the sources:

Article timeline

The articles in this story. Times are ET.

Oct 27
  1. Rohan PaulXPick
    Axiom's AI agent swarm formalizes the classification of finite simple groups in 24 days

    AIAxiom's AI agent swarm turned 15,000 pages across more than 500 papers into over 38 million lines of machine-verified code in 24 days. The effort also uncovered 34 serious discrepancies in the published record. The post quotes @biosemiote, who calls the formal proof of CFSG possibly among the largest open-source projects ever and says the work reflects decades of human mathematics. The attached chart shows Axiom's CFSG effort at 43.0M Lean lines, compared with other Lean formalizations since June 2025.

    Image from @rohanpaul_ai's post
  2. Rohan PaulXPick
    Axiom formalizes the Classification of Finite Simple Groups in Lean in 24 days

    AIAxiom has released a complete computer-checked Lean proof of the Classification of Finite Simple Groups, a roughly 15,000-page proof written by about 100 mathematicians. Over 24 days, engineers and mathematicians directed swarms of dozens of AI agents, which produced about 38 million lines of Lean code. Axiom credits the speed to rebuilding its models, harness and infrastructure together.

    Image from @rohanpaul_ai's post
  3. elvisXPick
    Elvis Saravia highlights Axiom's agent swarm formalizing the classification of finite simple groups

    AIElvis Saravia says Axiom used a swarm of AI agents to formally verify the Classification of Finite Simple Groups in Lean, about 50 million lines verified in 24 days. He says the agents claimed tasks, wrote and reviewed proofs, merged their own work through GitHub, and fixed hundreds of errata in the published proof.

    Image from @omarsar0's post
  4. Kevin Weil 🇺🇸XPick
    Axiom completes a full Lean formalization of the classification of finite simple groups

    AIKevin Weil says a full Lean formalization of the classification of finite simple groups is complete. He notes the original proof ran about 15,000 pages across 500 papers by 100 mathematicians, and says the formal version settles the result beyond doubt. The linked page on axiommath.ai describes the formalization.

  5. SantiagoX
    Axiom completes a Lean formalization of the Classification of Finite Simple Groups

    AIAxiom announces a complete Lean auto-formalization of the Classification of Finite Simple Groups, about 15,000 pages built by more than 100 mathematicians across generations. The author says multiple agents chose tasks, wrote proofs, audited each other's work, and merged their contributions, and adds that agents may eventually work independently for long periods.

Heat Trend

Current comparable heat
25
Peak comparable heat
39Oct 27, 12:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.