New proofs, counterexamples, bounds and formalizations reported by people and AI systems, newest first.

Reported developments, attributed to their sources. Inclusion is not an OpenMath assessment; evidence slots show what each source reports. Updated Oct 13, 2026.

October 2026 21

New proofNumber theoryopenai/math #022

The weak inhomogeneous Duffin–Schaeffer conjecture

Proves that for every real shift γ and finite-valued ψ:N→[0,∞)\psi:\mathbb N\to[0,\infty), divergence of ∑qϕ(q)ψ(q)/q\sum_q\phi(q)\psi(q)/q implies ∥qx−γ∥<ψ(q)\|qx-\gamma\|\lt \psi(q) for infinitely many q, for almost every x. Here ϕ is Euler's totient and the norm is distance to the nearest integer. Numerators are unrestricted; no monotonicity or Diophantine condition on γ is needed.

Unreleased OpenAI model
LFCR
Improved boundNumber theoryopenai/math #021

A quadratic bound for Jacobsthal’s function

Answers Jacobsthal's quadratic-bound question: every interval of Ck2Ck^2 consecutive integers contains an integer coprime to any prescribed positive integer with at most k distinct prime divisors, for an absolute constant C. The bound is uniform over prime sets and interval positions and removes the classical logarithmic loss.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #020

Squarefree quartics and power-free polynomial values

Proves the squarefree-values conjecture for irreducible integer quartics with no fixed prime-square divisor: squarefree values on positive integers have the predicted positive Euler-product density. More generally, establishes the (d−2)(d-2)-power-free density for irreducible integer polynomials of degrees four through eight under the necessary local condition; together with Browning's higher-degree theorem, this covers every d ≥ 4.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #019

The local p-adic section conjecture and global consequences

Proves that rational points on every smooth proper geometrically connected curve of genus at least two over a finite extension of ℚp correspond bijectively to conjugacy classes of sections of its full arithmetic étale fundamental group. It also proves Grothendieck's section conjecture over ℚ for the modular curves X0(N)X_0(N) and X1(N)X_1(N) of genus at least two.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #018

The Margulis–Platonov conjecture over global fields

Proves the Margulis–Platonov conjecture over every global field, including function fields of characteristic two. For an absolutely almost simple simply connected algebraic group G over k, every noncentral abstract normal subgroup of G(k)G(k) is the inverse image of an open normal subgroup in the finite product of its anisotropic nonarchimedean local groups.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #017

The irrationality exponent of π is 2

Proves that the irrationality exponent of π is exactly 2: for every ε > 0 and all sufficiently large denominators q, every rational p/qp/q satisfies ∣π−p/q∣≥q−2−ε|\pi-p/q|\ge q^{-2-\varepsilon}. This also proves convergence of the Flint–Hills series ∑n≥11/(n3sin⁡2n)\sum_{n\ge1}1/(n^3\sin^2 n), with angles in radians.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #016

Zilber–Pink in abelian varieties and the Siegel threefold

Proves the abelian Zilber–Pink conjecture over Q‾\overline{\mathbb Q}: every irreducible subvariety has finitely many maximal atypical subvarieties relative to its smallest containing torsion coset. It also proves the full curve case in the Siegel threefold A2\mathcal A_2 for Hodge-generic curves defined over Q‾\overline{\mathbb Q}, without boundary or reduction assumptions.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #015

Torus-packet equidistribution in prime, quartic, and sextic degrees

Proves Haar equidistribution without escape of mass for complete volume-weighted torus packets from totally real fields: arbitrary lattices and prescribed local types in fixed prime degree at least five, arbitrary-order Picard packets in primitive quartic fields, and maximal-order ideal-class packets in primitive sextic fields. Here primitive means having no proper intermediate field; the relevant order or field discriminant tends to infinity.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #014

Restricted geometric Langlands, global Arthur enhancements, and generic Ramanujan

Proves the restricted geometric Langlands equivalence for connected reductive groups on smooth projective connected curves over F‾q\overline{\mathbb F}_q under the four stated Lie-theoretic characteristic hypotheses. Over arbitrary algebraically closed fields of characteristic p > 0, the same conclusion holds assuming additionally that p is very good for the group and p∤∣WG∣p\nmid |W_G|. Over global function fields, proves Ramanujan at every place for globally generic cuspidal representations of split adjoint absolutely simple exceptional groups, without characteristic or ramification-depth restrictions, and at every unramified place for cuspidal representations of split adjoint absolutely simple groups with a generic unramified component. Assuming the finite-level Ramanujan–Arthur decomposition, constructs global Arthur enhancements of occurring cuspidal excursion parameters for split connected semisimple groups at full finite level, recovering the given parameters by diagonal specialization on the entire Weil group, including inertia.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #013

Ostmann’s inverse Goldbach conjecture

Proves that no finite modification of the primes can be written as A+BA+B with A,B⊆Z≥0A,B\subseteq\mathbb Z_{\ge0} each containing at least two elements. This resolves Ostmann's inverse Goldbach conjecture on additive indecomposability.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #012

Independent largest prime factors of consecutive integers

Resolves the Erdős–Pomerance joint Dickman conjecture: the logarithmic sizes of the largest prime factors of n and n+1n+1 are asymptotically independent in ordinary natural density. In particular, the integers satisfying P+(n)<P+(n+1)P^+(n)\lt P^+(n+1) have density 1/2.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #011

Prime-factor statistics of p−1p-1

Proves that the normalized ordered logarithms of the prime factors of p−1p-1, counted with multiplicity, converge jointly to the Poisson–Dirichlet law PD(1)\mathrm{PD}(1) as p ranges uniformly over primes up to x and x→∞x\to\infty. This resolves the Ford–Konyagin–Luca conjecture. It also proves that infinitely many integers n have more than n1−εn^{1-\varepsilon} totient preimages, for every ε > 0.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #010

Unrestricted pro-modularity at the prime two

Every continuous odd absolutely irreducible two-dimensional 2-adic representation of GQG_{\mathbb Q} unramified outside finitely many primes occurs in a completed Hecke algebra at some odd tame level. Also proves classical modularity up to Tate twist for irreducible odd representations with these finiteness conditions that are de Rham at 2 with distinct Hodge–Tate weights, resolving the dyadic Fontaine–Mazur case without residual restrictions.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #009

Function-field reconstruction from Milnor K-theory and Galois data

Reconstructs function fields of transcendence degree at least two over algebraically closed constants from K1M/ℓK^{\mathrm M}_1/\ell, K2M/ℓK^{\mathrm M}_2/\ell, and their product. These data recover the perfect closure and constants when ℓ differs from the characteristic, and the original field and its named base in equal characteristic. Also proves Bogomolov–Pop reconstruction from abelian-by-central pro-ℓ Galois data away from the characteristic.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #008

The Deligne–Drinfeld conjecture

Proves that the rational Grothendieck–Teichmüller Lie algebra, with the Ihara bracket, is freely generated by one element in each odd weight 3, 5, 7, …, resolving the Deligne–Drinfeld conjecture.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #007

Ordinary two-point correlations and the corrected Elliott conjecture

Proves the ordinary two-point Chowla conjecture, with a bound O(X/(log⁡X)c)O(X/(\log X)^c) for Liouville correlation sums along fixed nonproportional affine forms, where c > 0 is absolute. More generally, proves the binary corrected Elliott conjecture for complex multiplicative functions bounded by one when one factor is uniformly nonpretentious against each fixed Dirichlet character times nitn^{it} for ∣t∣≤X|t|\le X.

Unreleased OpenAI model
LFCR
New proofNumber theoryopenai/math #001

Milne’s rationality conjecture and algebraic specialization

Proves Milne's rationality conjecture for abelian varieties over Q‾\overline{\mathbb Q} with good reduction: specialized Hodge classes pair rationally with complementary divisor products, independently of cohomology theory. Together with result 032, every specialized Hodge class is represented by a single rational algebraic cycle simultaneously in all prime-to-p and crystalline realizations, for every residue characteristic p.

Unreleased OpenAI model
LFCR

September 2026 19

CounterexampleTheoretical computer science

Pinwheel scheduling kernel conjecture

A six-task counterexample is reported; exact periods and the reduction rule must be taken from the original witness before publishing.

GPT-6 Astra · Codex
LFCR
FormalizationNumber theory

Fermat's Last Theorem — new formalization

A new AI-assisted end-to-end Lean development is reported in September, formalizing an already proved theorem.

Claude · Tianyi Peng
LFCR