# OpenAI — Working with academic communities to build scientific understanding

- Company: OpenAI (openai.com)
- Announced: 2026-10-26T14:30:00+00:00
- Category: research-paper
- Coverage: 3 outlets
- Announcement: yes
- Group: covered
- Source: https://openai.com/index/building-scientific-understanding-with-academic-communities
- Record: https://forck.live/items/28002-working-with-academic-communities-to-build-scientific-understanding
- Subject: GPT / ChatGPT / API

OpenAI says its models produced three new results in mathematics, computer science and physics, and it is testing three different release strategies recommended by outside experts. The company describes the manuscripts as "a form of raw research data" and says it does not yet know which of the three approaches will prove most useful. For the mathematics result, the AI output is released so researchers can organize their own explanations; for the computer science result, an external group will explain it while the output stays public; for the physics result, an external group decides when materials are released. The collaborations follow four principles: transparency, no academic credit for OpenAI, recognition for the groups, and voluntary, unpaid, independent participation. In mathematics, the models produced a proof of the local smoothing conjecture in every dimension, which also resolves the Kakeya conjecture and Stein's restriction conjecture; the output was obtained on October 18th, 2026, formalized in Lean the next day, and published with a Lean proof in OpenAI's public GitHub repository. In computer science, the system proved that the Exponential Time Hypothesis is false, a result verified in Lean and now being shepherded by a group led by Russell Impagliazzo of UC San Diego. OpenAI notes that choosing initial shepherds shapes who gets involved first and creates asymmetry of information, and it welcomes feedback on broadening participation.

## Evidence

Verbatim from https://openai.com/index/building-scientific-understanding-with-academic-communities:

> OpenAI shares model results on the Kakeya conjecture, Exponential Time Hypothesis, and quantum chromodynamics—and how researchers can verify and build on them.

## Around this story

Outlets this record can name and link:

- Blockchain News — OpenAI shares new math and physics results with academic researchers: https://blockchain.news/news/openai-shares-new-math-and-physics-results-with-academic-researchers

## Follow-ups

- The Verge (reported, 26 Oct): OpenAI announces three big science discoveries and a new way to work with academics — https://forck.live/items/28034-openai-announces-three-big-science-discoveries-and-a-new-way-to-work-with
- Unite.AI (reported, 26 Oct): OpenAI Says Its Models Produced Three Proofs in Math, CS, and Physics — https://forck.live/items/28024-openai-says-its-models-produced-three-proofs-in-math-cs-and-physics

---

Record: https://forck.live/items/28002-working-with-academic-communities-to-build-scientific-understanding
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md