Blog
Disclosure, Unqualified Publication with Application and Lineage
Large Collaborative Proofs; The Classification of Finite Simple Groups and the Geometric Langlands Conjecture
Filed under MathematicsStatisticsLogicPhysicsScienceEngineeringPhilosophyLinguisticsTechniquesBibliography
Position
Publisher indexes for the quality of publications are important. Periodical stores and editorial archives are important. Neither should cause the inherent message of effort in authoring scientific or experimental work to be lost or dismissed, and prevent unqualified publishing. Peer review and editorial due diligence must continue, taking into account knowledge that was dismissed perhaps due to relevance. The message should nonetheless be applied to create progress through incremental outcomes, with immediate application where possible, and lineage in publishing and authorship is essential to this model.
Publisher indexes and editorial review remain important measures of quality. Our focus is incremental progress. This paper examines two large collaborative proofs: the Classification of Finite Simple Groups and the proof of the geometric Langlands conjecture. Both were applied well before formal editorial closure.
Case One: The Classification of Finite Simple Groups
Statement
Every finite simple group is isomorphic to exactly one of the following:
- a cyclic group of prime order;
- an alternating group of degree at least 5;
- a group of Lie type, drawn from 16 infinite families;
- one of 26 sporadic groups.
A simple group has no non-trivial proper normal subgroups. The Jordan–Hölder theorem shows that every finite group decomposes into simple factors. The classification therefore describes the building blocks of all finite groups.
Scale
The proof is the largest collaborative proof in mathematics. It occupies roughly 10,000 to 15,000 journal pages. It is spread across about 500 articles by over 100 authors.
Milestones
- 1955
- The Brauer–Fowler theorem links simple groups to centralisers of involutions.
- 1963
- The Feit–Thompson theorem proves that every group of odd order is solvable [1].
- 1965–1975
- Most sporadic groups are discovered; the Monster is constructed by Griess in 1982.
- 1983
- Gorenstein announces completion [7].
- 1994
- Gorenstein, Lyons and Solomon begin publishing a consolidated second-generation proof [4].
- 2004
- Aschbacher and Smith close the quasithin gap [2, 3, 9].
- 2012
- The Feit–Thompson theorem is formally verified in the Coq proof assistant [11].
Case Two: The Geometric Langlands Conjecture
Statement
Let X be a smooth projective curve over a field of characteristic zero. Let G be a reductive group with Langlands dual Ǧ. The proven statement is a categorical equivalence between two categories:
- the automorphic side: D-modules on BunG, the moduli stack of G-bundles on X;
- the spectral side: ind-coherent sheaves with nilpotent singular support on LocSysǦ, the moduli stack of Ǧ-local systems on X.
The proof applies in the de Rham setting.
Scale and Authorship
The proof runs to roughly 800 to 1,000 pages across five papers posted in 2024 [12–16]. Nine mathematicians contributed: D. Arinkin, D. Beraldo, J. Campbell, L. Chen, J. Faergeman, D. Gaitsgory, K. Lin, S. Raskin and N. Rozenblyum. The work builds on about thirty years of foundational work in derived algebraic geometry [18–21]. The conjecture was formulated in the 1980s, building on the programme initiated by Langlands in 1967 [22, 23]. Gaitsgory received the 2025 Breakthrough Prize in Mathematics in recognition of this programme.
Discussion
Evidence for Early Application
The classification was used as a working theorem from 1983 onward. Results such as the classification of 2-transitive permutation groups and the O’Nan–Scott framework rest on it. Waiting until 2004 would have frozen two decades of progress in permutation groups, combinatorics and model theory.
The geometric Langlands proof circulated on arXiv and on the author’s institutional page from 2024 [17]. Researchers built on it immediately. Its value did not wait on journal acceptance.
Perelman’s proof of the Poincaré conjecture is the extreme case. It was never submitted to a journal. The community nonetheless verified it, applied it, and awarded both the Fields Medal and the Millennium Prize.
The Cost of Unqualified Application
The quasithin gap is the counterexample to unqualified early application. For twenty years, every result citing the classification depended on an incomplete argument. Most users did not know this. The gap closed successfully, but the risk was real and largely invisible to those carrying it.
The discipline that emerged is conditional application. Careful authors stated “assuming CFSG” in their hypotheses. That convention allowed progress while keeping the dependency traceable. It is the practical answer: apply immediately, but declare what the application rests on.
Lineage
The second-generation classification proof exists because the knowledge was concentrated in an ageing group of specialists. Gorenstein, Lyons and Solomon recognised that the proof could become effectively unrecoverable if the lineage broke [4, 8]. The 2026 priority dispute between Buckmaster, Alpöge and OpenAI over the Navier–Stokes result shows the same issue in a new form [24]. Authorship and timestamped disclosure are now the primary mechanism for establishing who contributed what.
Formal Verification
Formal verification separates correctness from editorial process. The 2026 Navier–Stokes and Fermat formalisations were accepted as correct because the Lean code compiles [25, 27]. Correctness can therefore be established quickly and mechanically. Peer review can concentrate on what machines cannot judge: significance, readability, and relevance to the question that matters. The Navier–Stokes case shows this split clearly. The proof is accepted as valid, yet reviewers still ask whether it solved the right problem [26].
Proposed Model
- Disclose early, with clear authorship and timestamps.
- Verify formally where possible.
- Apply conditionally, with declared dependencies.
- Let editorial review assess significance on its own timeline.
References
- W. Feit and J. G. Thompson. Solvability of groups of odd order. Pacific Journal of Mathematics, 13(3):775–1029, 1963.
- M. Aschbacher and S. D. Smith. The Classification of Quasithin Groups I: Structure of Strongly Quasithin K-groups. Mathematical Surveys and Monographs 111. American Mathematical Society, 2004.
- M. Aschbacher and S. D. Smith. The Classification of Quasithin Groups II: Main Theorems. Mathematical Surveys and Monographs 112. American Mathematical Society, 2004.
- D. Gorenstein, R. Lyons and R. Solomon. The Classification of the Finite Simple Groups, Numbers 1 onward. Mathematical Surveys and Monographs 40. American Mathematical Society, 1994–.
- M. Aschbacher, R. Lyons, S. D. Smith and R. Solomon. The Classification of Finite Simple Groups: Groups of Characteristic 2 Type. Mathematical Surveys and Monographs 172. American Mathematical Society, 2011.
- D. Gorenstein. Finite Simple Groups: An Introduction to Their Classification. Plenum Press, 1982.
- D. Gorenstein. The Classification of Finite Simple Groups, Volume 1: Groups of Noncharacteristic 2 Type. Plenum Press, 1983.
- R. Solomon. A brief history of the classification of the finite simple groups. Bulletin of the American Mathematical Society, 38(3):315–352, 2001.
- M. Aschbacher. The status of the classification of the finite simple groups. Notices of the American Mathematical Society, 51(7):736–740, 2004.
- J. H. Conway, R. T. Curtis, S. P. Norton, R. A. Parker and R. A. Wilson. ATLAS of Finite Groups. Oxford University Press, 1985.
- G. Gonthier et al. A machine-checked proof of the Odd Order Theorem. In Interactive Theorem Proving (ITP 2013), Lecture Notes in Computer Science 7998, pages 163–179. Springer, 2013.
- D. Gaitsgory and S. Raskin. Proof of the geometric Langlands conjecture I: construction of the functor. arXiv:2405.03599, 2024.
- D. Arinkin, D. Beraldo, J. Campbell, L. Chen, J. Faergeman, D. Gaitsgory, K. Lin, S. Raskin and N. Rozenblyum. Proof of the geometric Langlands conjecture II: Kac–Moody localization and the FLE. arXiv:2405.03648, 2024.
- J. Campbell, L. Chen, J. Faergeman, D. Gaitsgory, K. Lin, S. Raskin and N. Rozenblyum. Proof of the geometric Langlands conjecture III: compatibility with parabolic induction. arXiv:2409.07051, 2024.
- D. Arinkin, D. Beraldo, L. Chen, J. Faergeman, D. Gaitsgory, K. Lin, S. Raskin and N. Rozenblyum. Proof of the geometric Langlands conjecture IV: ambidexterity. arXiv:2409.08670, 2024.
- D. Gaitsgory and S. Raskin. Proof of the geometric Langlands conjecture V: the multiplicity one theorem. arXiv:2409.09856, 2024.
- D. Gaitsgory. Geometric Langlands conjecture: proof. https://people.mpim-bonn.mpg.de/gaitsgde/GLC/.
- D. Arinkin and D. Gaitsgory. Singular support of coherent sheaves and the geometric Langlands conjecture. Selecta Mathematica, 21(1):1–199, 2015.
- D. Gaitsgory. Outline of the proof of the geometric Langlands conjecture for GL(2). Astérisque, 370:1–112, 2015.
- D. Gaitsgory and N. Rozenblyum. A Study in Derived Algebraic Geometry, Volumes I and II. Mathematical Surveys and Monographs 221. American Mathematical Society, 2017.
- D. Arinkin, D. Gaitsgory, D. Kazhdan, S. Raskin, N. Rozenblyum and Y. Varshavsky. The stack of local systems with restricted variation and geometric Langlands theory with nilpotent singular support. arXiv:2010.01906, 2020.
- E. Frenkel. Lectures on the Langlands program and conformal field theory. In Frontiers in Number Theory, Physics, and Geometry II, pages 387–533. Springer, 2007.
- A. Beilinson and V. Drinfeld. Quantization of Hitchin’s integrable system and Hecke eigensheaves. Unpublished manuscript, 1991.
- Quanta Magazine. AI has solved one of math’s $1 million Millennium Prize problems. 8 September 2026. quantamagazine.org.
- NPR. Mathematicians learn little from AI completing unsolved problem. 22 September 2026. npr.org.
- Scientific American. Did OpenAI solve the wrong Navier–Stokes problem? September 2026. scientificamerican.com.
- Decrypt. AI just solved a 350-year-old math problem by writing the longest proof ever. 5 September 2026. decrypt.co.