# The Margulis–Platonov conjecture over global fields > Normal subgroups of these simple groups over any global field are claimed to come from finitely many local pieces. Plain-English explainer with sources. Canonical: https://closedform.nishachay.tech/e/openai-math-2026-018/ Last updated: 2026-10-14 **Normal subgroups of these simple groups over any global field are claimed to come from finitely many local pieces.** **Trust:** Claimed, not checked · **Papers:** 2 · **First paper:** September 23, 2026 · **Field:** Number theory · **Entry:** 018 *Explained by AI.* Drafted by one AI model and checked sentence by sentence against the paper by a different one. Spot a mistake? Report an error. Take a matrix group such as the 2 × 2 matrices of determinant 1 with fraction entries. A normal subgroup is a part of it that conjugation never moves outside itself. The Margulis–Platonov conjecture says such parts come only from finitely many compact local groups. Two papers claim to prove it: one over number fields, one over function fields, including characteristic two. Many cases were already known; these papers claim the remaining ones. ## At a glance - **First step:** 1979 (G. A. Margulis, who formulated the description by anisotropic local factors) - **Significance:** Major. It claims the last open types of a long-standing conjecture on normal subgroups of arithmetic groups. - **Trust:** Claimed, not checked - **Field:** Number theory (2 papers) ## The explanation Take the 2 × 2 matrices with fraction entries and determinant 1. The matrix with rows (2, 1) and (1, 1) is one. Its determinant is 2 × 1 − 1 × 1 = 1. Together these matrices form a group called SL₂ over the rationals. A normal subgroup is a part that stays inside itself when you conjugate by any member. The center here is just the identity and its negative. For this group, any normal subgroup bigger than the center is the whole group. Not every such group is that rigid. Now take quaternions a + bi + cj + dk with fraction coefficients. Here i, j and k are three square roots of −1 with ij = k. Keep those with a² + b² + c² + d² = 1. One example is 3/5 + 4/5 i, since 9/25 + 16/25 = 1. These form a group too, but it can have many normal subgroups. The reason is local. Viewed with 2-adic numbers (a way of measuring numbers by divisibility by 2), this group becomes compact (closed and bounded). Compact local groups have many open normal subgroups, and these pull back to the rational group. The Margulis–Platonov conjecture says that is the only source. Take a group G of the kind studied, called absolutely almost simple and simply connected. Collect the finitely many places where G becomes anisotropic, so its local group is compact[1]. Then every normal subgroup not inside the center comes from an open normal subgroup of the product of those compact groups[2]. When there are no such places, the only option is the whole group[3]. The normal subgroups are abstract: no extra condition is put on them[4]. The conjecture grew from a question of Kneser on quaternions and a proposal of Platonov. Margulis formulated it in 1979[5]. Many cases followed. Groups with no compact local places were handled through the Kneser–Tits theorem[6]. Chernousov treated many anisotropic groups[7]. For inner type A, built from division algebras, Segev and Segev–Seitz completed the case[8]. Rapinchuk, Segev and Seitz later proved it over all global fields for that type[9]. The open cases were certain outer type A, triality D₄ and E₆ groups over number fields[10]. The number-field paper claims to settle those remaining cases[2,10]. The function-field paper claims the conjecture over every global function field, a field built from polynomials over a finite field[11]. That includes characteristic two. There, earlier results had already reduced the problem to special unitary groups of outer type A[12]. The hard step is a finite obstruction. Known tools show a normal subgroup has finite index. They also show its closure matches the local picture. What remains is a finite gap between the abstract subgroup and that closure[13]. For the special unitary groups, the paper rules out a finite simple group as the quotient. It colors each rational element by its image, then argues that a translation-invariant random coloring leads to a contradiction. The number-field paper also gives a consequence for the congruence subgroup property. Feeding the theorem into a 2026 theorem of Rapinchuk gives that property when the set S of places has positive density[14]. It does not address Serre’s conjecture for finite S[15]. How sure is this? No Lean formalization is listed for these papers in the openai/math release. So no computer proof checker has checked them. The proofs rely on deep results, including the classification of finite simple groups. No specialist review has been published. **What it doesn't show.** This does not settle Serre's congruence subgroup conjecture in general, only a positive-density case that follows from it. ## Where it came from This result comes from an unreleased OpenAI model, published in the openai/math release on 6 October 2026. ## Hype check Major only if both papers hold. No Lean formalization is listed and no specialist review is published. Many cases were proved by others; these papers claim the remaining ones. ## Words used - **Normal subgroup:** A part of a group that conjugation by any member keeps inside itself. - **Center:** The members of a group that commute with every member. - **Anisotropic:** For these groups, having a compact local group, with no unipotent elements to move things far. - **Compact:** Closed and bounded, so that every sequence has a limit point inside. - **p-adic numbers:** A number system that measures size by divisibility by a prime p. - **Global field:** A number field, or a field of rational functions over a finite field and its finite extensions. - **Congruence subgroup property:** The property that every finite-index normal subgroup contains a subgroup defined by congruences. ## The original papers (2) - [The Margulis–Platonov conjecture over global function fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-global-function-fields-October-5-2026/margulis-platonov-global-function-fields.pdf) - October 5, 2026 - [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf) - September 23, 2026 ## Quoted lines from the paper (15) [1] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > The set $A$ is finite. The product $H_A$ is given its product topology; if $A$ is empty, it is the trivial group. [2] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > Let $k$ be a number field and $G$ an absolutely almost simple simply connected algebraic $k$-group. If $N\triangleleft G(k)$ is not contained in $Z(G(k))$, then there is an open normal subgroup $W\triangleleft H_A$ such that \[ N=\delta_A^{-1}(W). \] [3] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > In the case $A=\varnothing$, the conclusion is $N=G(k)$, not merely that $N$ has finite index or is dense in a local topology. [4] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > The Margulis--Platonov conjecture asserts that the noncentral normal subgroups of $G(k)$ are accounted for by this finite collection of compact local groups. [5] From [The Margulis–Platonov conjecture over global function fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-global-function-fields-October-5-2026/margulis-platonov-global-function-fields.pdf): > Margulis formulated the description by anisotropic local factors over global fields in. [6] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > The isotropic case follows from the Kneser--Tits theorem over global fields; Gille's account includes the final outer-$E_6$ cases, using work of Garibaldi and Chernousov--Timoshenko. [7] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > For anisotropic groups, Chernousov proved projective simplicity of the rational points when the absolute rank is at least two and the group splits over a quadratic extension. [8] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > Segev obtained a commuting-graph obstruction, and Segev--Seitz verified the required finite-simple-group properties, completing this case. [9] From [The Margulis–Platonov conjecture over global function fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-global-function-fields-October-5-2026/margulis-platonov-global-function-fields.pdf): > Rapinchuk--Segev--Seitz proved the stronger result that every finite quotient of the multiplicative group of a finite-dimensional division algebra is solvable; their Theorem 6.4 proves the Margulis--Platonov conjecture for inner type $A$ over arbitrary global fields. [10] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > They include isotropic groups and all inner forms of type $A$, as well as the other types outside the remaining anisotropic outer $A$, trialitarian $D_4$, and $E_6$ cases. [11] From [The Margulis–Platonov conjecture over global function fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-global-function-fields-October-5-2026/margulis-platonov-global-function-fields.pdf): > This holds in every positive characteristic, including when $G$ is anisotropic over $k$. [12] From [The Margulis–Platonov conjecture over global function fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-global-function-fields-October-5-2026/margulis-platonov-global-function-fields.pdf): > Together with the established inner-type-$A$ theorem, these results reduce the present problem to anisotropic outer forms of type $A$, represented by special unitary groups. [13] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > This is the distinction between the desired abstract equality and the closure statement furnished by approximation. [14] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > Theorem … supplies the Margulis--Platonov hypothesis in Rapinchuk's positive-density theorem, giving the following consequence. [15] From [The Margulis–Platonov conjecture over number fields](https://github.com/openai/math/blob/main/preprints/The-Margulis-Platonov-conjecture-over-number-fields-September-23-2026/paper.pdf): > The corollary is restricted to the stated positive-density range: it does not address Serre's finite-$S$ conjecture, arbitrary infinite $S$, or centrality or finiteness of congruence kernels outside that range. ## Other sources - [Y. Segev, “On finite homomorphic images of the multiplicative group of a division algebra”, Annals of Mathematics 149 (1999), 219–251](https://doi.org/10.2307/121024) - [Y. Segev and G. M. Seitz, “Anisotropic groups of type A_n and the commuting graph of finite simple groups”, Pacific Journal of Mathematics 202 (2002), 125–225](https://doi.org/10.2140/pjm.2002.202.125) - [A. S. Rapinchuk, Y. Segev and G. M. Seitz, “Finite quotients of the multiplicative group of a finite dimensional division algebra are solvable”, Journal of the AMS 15 (2002), 929–978](https://doi.org/10.1090/S0894-0347-02-00393-4) - [A. S. Rapinchuk, “The congruence subgroup property for S-arithmetic subgroups of simple algebraic groups when S has positive Dirichlet density”, arXiv:2603.27472 (2026)](https://arxiv.org/abs/2603.27472v1) - [openai/math release notes (2026): model provenance and review status](https://github.com/openai/math/blob/main/README.md): “manuscripts and supporting proof artifacts produced by an internal OpenAI model” ## Related - [Goldfeld’s conjecture: densities and mean analytic rank](https://closedform.nishachay.tech/e/openai-math-2026-006/) - Number theory - [The local p -adic section conjecture and global consequences](https://closedform.nishachay.tech/e/openai-math-2026-019/) - Number theory - [The weak inhomogeneous Duffin–Schaeffer conjecture](https://closedform.nishachay.tech/e/openai-math-2026-022/) - Number theory - [Ostmann’s inverse Goldbach conjecture](https://closedform.nishachay.tech/e/openai-math-2026-013/) - Number theory - [The Deligne–Drinfeld conjecture](https://closedform.nishachay.tech/e/openai-math-2026-008/) - Number theory - [Number theory](https://closedform.nishachay.tech/science/mathematics/number-theory/) - Field · 31 results - [Named conjectures](https://closedform.nishachay.tech/topic/named-conjectures/) - Topic · 85 results - [Open since the 1970s](https://closedform.nishachay.tech/open-since/1970s/) - Era · 63 results - [Claimed, not yet checked](https://closedform.nishachay.tech/trust/claimed/) - Trust · 130 results - [Report an error](https://github.com/nishachay/closed-form/issues/new?template=report-error.yml&labels=entry-error&title=%5Bentry%20018%5D%20)