AxiomProver

Headlines · 3

1. 4d
   Axiom reports a 24-day Lean formalization of CFSG

   Axiom Math says that dozens of AI agents, coordinated with human mathematicians and engineers, completed a Lean formalization of the first-generation proof of the Classification of Finite Simple Groups in 24 days. The company reports about 43.1 million lines of Lean code and 2.05 million declarations, and has opened the repository for inspection. Axiom also says the formalization uncovered an issue in a published argument for the Unbalanced Group Theorem, leading to an alternative route and the result “Preservation of simple tops.” The repository’s review status is described as a team assessment; external mathematicians still need to verify that the formal statement matches CFSG and independently assess the claimed issue and repair.

   Hacker News 首页（100+ 分） · 🔥 10
2. 7d
   OpenAI Releases Three Preliminary AI Research Findings

   OpenAI announced three preliminary findings spanning mathematics, computer science, and physics, presenting the manuscripts as raw research material that requires external verification and interpretation. The release follows its publication of 722 AI-generated mathematics manuscripts, drawing criticism from Terence Tao, who called parts of the resulting proof workload “toxic waste” and warned that corporate competition was neglecting human understanding. OpenAI denied claims by Tao that it had prioritized the work at the highest level, paused new-user registration, or spent tens of millions of dollars on the effort.

   钛媒体 · 🔥 10
3. Oct 18
   After OpenAI's quasi-Riemann claim, mathematicians push number-theory results

   According to a 36Kr repost of a New Zhiyuan article, OpenAI claimed on October 6 that its AI proved a quasi-Riemann hypothesis, excluding Landau–Siegel zeros, drawing positive comments from number theorist Hector Pasten and others. The report says Cureton and Thorner posted a Hexagon preprint on October 12 applying the tools to finite-order Hecke L-functions to obtain a zero-free region to the right of 1-1/6000, which an open-source community then pushed to 1/32, with a Lean formalization generated by the math agent AxiomProver. It adds that a preprint by Frei, Somos and Widmer on arXiv used these results to prove unconditionally the 2007 Ellenberg–Venkatesh bound on ℓ-torsion in class groups, and that the paper states no artificial intelligence tools were used. cnBeta's republished headline says AI again broke the generalized Riemann hypothesis, but the described work concerns zero-free regions rather than the full GRH.

   36氪 人工智能 · 🔥 0

Experience and discussion from the community

Share my AxiomProver experienceAsk about AxiomProver