---
id: "2026-10-09-arxiv-ai-disclosed-results-early-october-2026"
url: "https://postcutoff.com/e/2026-10-09-arxiv-ai-disclosed-results-early-october-2026/"
as_of: "2026-10-12T07:22:00+02:00"
date: "2026-10-09"
date_precision: day
category: science
importance: 3
confidence: medium
status: [Awaiting review]
verification: Preprint only
sources: 14
editor: Adam Bicz
human_review: null
version: null
---

As of: 2026-10-12 07:22 CEST.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code). Human editor: Adam Bicz.
Canonical page: https://postcutoff.com/e/2026-10-09-arxiv-ai-disclosed-results-early-october-2026/

# More conjectures settled with disclosed AI help, next to human-only results

Full title: arXiv, 5–9 October 2026: more conjectures settled with disclosed AI help (Quillen's local parameter conjecture in dimension 4, Laughlin gap, poset reconstruction), next to human-only results

A catalogue of smaller arXiv preprints from 5–9 Oct 2026 that claim to settle a named conjecture or open problem, with the AI role as the authors state it. AI-developed proofs: Quillen's local parameter conjecture in dimension four ("ChatGPT and Codex (GPT-6) were used to develop the proof"), reconstruction of finite posets of width three ("discovered by ChatGPT 6 Pro"), and a much better lower bound on the Laughlin gap that improves an OpenAI release result ("completed by the AI model Claude Opus 5.5"). Others used AI for literature search or checking, or state none. All are unrefereed. Larger results of the same week have their own entries.

## Key facts
- Quillen's local parameter conjecture, dimension four (Yunpeng Li, CUHK Shenzhen, arXiv 2610.13469, 72 pages): for a regular local ring R of mixed characteristic and dimension four and a regular parameter r, every finitely generated projective module over R[1/r] is free. Known before when R contains a field (Bhatwadekar–Rao, Popescu) and in dimension ≤ 3 (Gabber). Disclosure: 'OpenAI's ChatGPT and Codex (GPT-6) were used to develop the proof and prepare the manuscript. The human author takes full responsibility for the mathematical arguments and the final manuscript.' Single author, not expert-checked
- Laughlin gap (Qiang Gao, arXiv 2610.12700, 20 pages): raises the constant in OpenAI's Fock-space inequality from 0.0461673… to 0.2190037845…, 'certified by exact rational arithmetic', so the spectral gap above the Laughlin state at filling 1/3 is at least 1/5 instead of 1/25 for large spheres (numerics put the true gap near 0.4). 'The proof of the improvement and the SDP search were completed by the AI model Claude Opus 5.5 (Anthropic) under the author's direction'; the author noticed that OpenAI's certificate 'appeared to be hand-picked rather than obtained by optimization' and asked Claude to set up a semidefinite-programming search. A Lean proof by 'ChatGPT 6.1 Sol' is 'in progress'
- Reconstruction of finite posets of width three (Patrizio Cintioli, University of Camerino, arXiv 2610.12690, 19 pages): every finite ordered set of width at most three and at least four elements is reconstructible from its deck, the width-three part of an open problem of Schröder. 'The main result was discovered by ChatGPT 6 Pro (OpenAI), which was also used in developing the proof and preparing the manuscript. The mathematical proofs were also subjected to an independent adversarial review using Claude Fable 5.1 (Anthropic).'
- Tate conjecture for ordinary Calabi–Yau threefolds (Lukas Brantner, Oxford and CNRS, arXiv 2610.13939, 6 pages): integral geometric Tate conjecture over finite fields via canonical lifts. AI as a literature guide: 'We devised the overall proof strategy. ChatGPT was used to explore the literature and review earlier drafts. The model directed our attention to the work of Patakfalvi and Zdanowicz, which helped us handle the case of divisor classes when b1(X) > 0.'
- Randomized Frank–Wolfe lower bound (Sebastian Pokutta, Zuse Institute Berlin, arXiv 2610.13590): an Ω(L·diam(S)²/T²) lower bound for randomized linear-minimization-oracle methods over strongly convex sets. 'Generative AI tools assisted throughout the research process with formulation and revision of mathematical arguments, implementation of verification and plotting scripts, formalization, and manuscript editing', with 'machine-checked proofs for the results covered by the formalization'
- Light AI use: the biwheel conjecture for braces (Feng and Zhou, Jimei University, arXiv 2610.12955; de Carvalho–Lucchesi–Murty's conjecture that the minimum number of perfect matchings of a brace of order n is (n − 2)²/4; AI tools used 'to assist in discussing proof strategies, checking proofs, and improving the exposition'); counterexamples to Minkowski-type inequalities for modules (Jorge Pérez, arXiv 2610.13888; 'the vast majority of its theoretical results were obtained without the aid of generative AI tools', calculations with GPT-5.6 Sol and Codex)
- A paper about an AI proof: Farzad Aryan's 'On the Imaginary Part of ζ′/ζ' (arXiv 2610.12853) is 'motivated by the recent breakthrough of Anthropic's Claude' on zeta zeros and includes 'An Expository Note on Claude's Argument', explaining how Claude 'replaced the termwise positivity argument by a global linear-algebraic analysis'. Aryan is the author of the 2017 unconditional pair-correlation framework that the Claude result built on
- Same week, no AI disclosure found in the PDF (sweep scan): Pisanski's 1980 conjecture on quadrilateral embeddings of regular bipartite graphs (Metzger and Ulrigg, University of Washington, 2610.13085); counterexamples to the Ambro–Kawamata effective non-vanishing conjecture (Rahul Ajit, 2610.09270); a counterexample to Oxley's circuit–cocircuit intersection conjecture from the extended binary Golay code (Houshan Fu, 2610.06033); the fractional Lane–Emden conjecture (Mei and Wei, 2610.10188); the Mueller–Ho conjecture (2002) on vortex lattices in two-component Bose–Einstein condensates (Luo and Wei, 2610.12203)
- None of these results had press coverage, a Hacker News thread or a public expert assessment that we could find on 12 Oct 2026

## What happened

The stream of AI-assisted arXiv results described in the summer 2026 catalogue continued in the first full week of October, the
week of OpenAI's 722-manuscript release. This entry lists the smaller items of 5–9 Oct. The larger ones have their own entries:
the Lean formalization of geometrization, the counterexample to Nirenberg's conjecture, near-linear Subset Sum, universality of
random permutations, explicit tensors of border rank 3n and the holomorphic cancellation problem.

Three patterns are visible.

- **Single authors with AI-developed proofs of hard statements.** The dimension-four case of Quillen's local parameter conjecture
  is a 72-page proof in algebraic K-theory by one author from a data-science school, developed with ChatGPT and Codex. Nobody
  else has checked it yet, so confidence is low until specialists comment.
- **Outsiders improving OpenAI's results within days.** Qiang Gao read OpenAI's Laughlin-gap proof, saw that its certificate was
  not optimized, and had Claude Opus 5.5 run a semidefinite-programming search that multiplied the constant by almost five.
- **Honest small roles.** Brantner credits ChatGPT with pointing to one relevant paper. Several authors state that the results
  predate AI tools or used them only for checking and wording.

Human-only papers settling named conjectures appeared in the same week.

## Why it matters

The catalogue shows what the everyday use of frontier models in research mathematics looked like in October 2026: routine,
disclosed in a sentence or two, and uneven in reliability. Results proved by a model and checked by one person need independent
confirmation before they are cited as theorems.

## What is disputed or not yet verified

- Verification: Unrefereed preprints; AI role self-reported by authors

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): 162 days after its cutoff
- Claude Opus 5.5 (training cutoff June 2026): 101 days after its cutoff
- Gemini 3.8 Flash (training cutoff March 2026): 192 days after its cutoff
- Grok 4.7 (training cutoff May 2026): 131 days after its cutoff

## Sources

1. [arXiv 2610.13469: Quillen's local parameter conjecture in dimension four](https://arxiv.org/abs/2610.13469) (arxiv.org, paper)
2. [arXiv 2610.12700: Improving the lower bound on the Laughlin gap: An SDP search](https://arxiv.org/abs/2610.12700) (arxiv.org, paper)
3. [OpenAI math release: A Fock-space inequality and the Laughlin spectral gap (Lean)](https://github.com/openai/math/tree/main/lean/OAI/Analysis/LaughlinFock) (github.com, code)
4. [arXiv 2610.12690: Reconstruction of Finite Posets of Width Three](https://arxiv.org/abs/2610.12690) (arxiv.org, paper)
5. [arXiv 2610.13939: The Tate Conjecture for Ordinary Calabi-Yau Threefolds](https://arxiv.org/abs/2610.13939) (arxiv.org, paper)
6. [arXiv 2610.13590: Lower Bounds for Randomized Linear-Oracle Optimization over Strongly Convex Sets](https://arxiv.org/abs/2610.13590) (arxiv.org, paper)
7. [arXiv 2610.12955: The biwheel conjecture for braces](https://arxiv.org/abs/2610.12955) (arxiv.org, paper)
8. [arXiv 2610.13888: Counterexamples to Minkowski-type inequalities and the Rees converse for modules](https://arxiv.org/abs/2610.13888) (arxiv.org, paper)
9. [arXiv 2610.12853: On the Imaginary Part of ζ′/ζ](https://arxiv.org/abs/2610.12853) (arxiv.org, paper)
10. [arXiv 2610.13085: Resolving Pisanski's Conjecture](https://arxiv.org/abs/2610.13085) (arxiv.org, paper)
11. [arXiv 2610.09270: Counterexamples to the Ambro–Kawamata effective non-vanishing conjecture](https://arxiv.org/abs/2610.09270) (arxiv.org, paper)
12. [arXiv 2610.06033: A counterexample to the circuit-cocircuit intersection conjecture](https://arxiv.org/abs/2610.06033) (arxiv.org, paper)
13. [arXiv 2610.10188: A proof of the fractional Lane–Emden conjecture](https://arxiv.org/abs/2610.10188) (arxiv.org, paper)
14. [arXiv 2610.12203: A Complete Proof of Mueller–Ho Conjecture](https://arxiv.org/abs/2610.12203) (arxiv.org, paper)

## Changes

- 2026-10-12 (filed): Created from the PDFs listed by the 2026-10-12 sweep (two days of arXiv listings)

## Related

- 2026-10-09: [Thurston's geometrization conjecture formalised in Lean](https://postcutoff.com/e/2026-10-09-geometrization-conjecture-lean-formalization/index.md)
- 2026-10-09: [Nirenberg's 1962 conjecture on closed asymptotic curves disproved](https://postcutoff.com/e/2026-10-09-nirenberg-conjecture-closed-asymptotic-curves-counterexample/index.md)
- 2026-10-09: [Subset Sum in O(n + N^(1+ε)) time claimed](https://postcutoff.com/e/2026-10-09-subset-sum-near-linear-time-gpt-astra-ultra/index.md)
- 2026-10-09: [Random permutations of quadratic length contain every pattern](https://postcutoff.com/e/2026-10-09-random-permutations-universal-alon-gpt-6-pro/index.md)
- 2026-10-08: [Holomorphic Cancellation Problem (1998) answered in the negative, conditional on a newly announced exotic affine space; proofs developed with generative AI (Gaofeng Huang, Harvard)](https://postcutoff.com/e/2026-10-08-holomorphic-cancellation-problem-counterexample/index.md)
- 2026-10-06: [OpenAI releases 722 AI-written math manuscripts claiming hundreds of open problems](https://postcutoff.com/e/2026-10-06-openai-math-release-722-manuscripts/index.md)
- 2026-10-06: [Explicit tensors of border rank 3n − o(n), past the 'linear flattening barrier'](https://postcutoff.com/e/2026-10-06-explicit-tensors-border-rank-3n-gpt-6-astra/index.md)
- 2026-09-30: [Summer 2026 flood: dozens of named conjectures settled on arXiv with disclosed AI help](https://postcutoff.com/e/2026-09-30-ai-assisted-conjecture-wave-summer-2026/index.md)
- 2026-08-10: [Claude proves more than two-thirds of Riemann zeta zeros are simple and on the critical line (up from 41.6%)](https://postcutoff.com/e/2026-08-10-claude-riemann-zeta-zeros-two-thirds/index.md)