OpenAI model allegedly disproves Exponential Time Hypothesis
OpenAI claims an internal model disproved the Exponential Time Hypothesis (ETH), a long-standing conjecture in computer science. The company announced the result on October 26, saying its model found the disproof on October 16. OpenAI verified the claim using Lean, a proof-verification language, and published the model’s write-up and proof on GitHub.
Key points
- OpenAI’s internal model allegedly disproves the Exponential Time Hypothesis (ETH) on October 16, verified via Lean proof
- Researchers from UC San Diego and MIT will independently review the claim, but stress no endorsement or payment
- OpenAI released raw data and agreed to fund a conference for further scrutiny
Three researchers—Russell Impagliazzo, Ramamohan Paturi, and Ryan Williams—will independently review the claim. They emphasized their role is not an endorsement and warned that the result requires careful scrutiny. OpenAI agreed to release raw data and fund a conference for further examination. The researchers also raised questions about how credit should be assigned when AI discovers results first, referencing concerns from the Leiden declaration in June.
OpenAI says its AI model has disproved the Exponential Time Hypothesis
thenextweb.com · 26 October 2026
Loading the full article…
This text was published by thenextweb.com and written by Ana Maria Constantin. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Anthropic adds self-assessment layer to Claude safety monitoring · 2 src
- IEEE launches AI Research Suite for Xplore Digital Library · 1 src
- H2O.ai’s TimeH2O-v1 tops time-series benchmarks with 91.9% win rate · 1 src
- Foundation models match but sometimes lag task-specific models in prostate pathology · 1 src
- Researchers release ProBEiT, a 400M-parameter protein model · 1 src
Comments
via GitHub Discussions