Grok 4.5 Potentially Provides Original Math Counterexample
LiorOnAI · x · 2026-07-11
Reports indicate that Grok 4.5 provided a new counterexample to an open mathematical problem. If independently verified, it would suggest the model isn't merely regurgitating existing work but driving an original mathematical discovery.
The post explains that the problem relates to hypercontractivity, involving harmonic analysis, probability, and partial differential equations. The previously known conclusion was that the property holds in 3 dimensions and fails by 13 dimensions, but the exact dimension of the "first failure" was unclear. The counterexample constructed by Grok allegedly pushes this critical point to 4 dimensions, thereby refining previous results.
Related event: Grok 4.5 Reportedly Discovers Research-Level Math Counterexample(5 posts)→
More from Models
- Together AI sits down with Meta to explain Muse Spark Open — togethercompute · 2026-10-27
- GPT-6.1 Terra hits ~60% on Agents' Last Exam at $1/task, 9x cheaper than Opus 5.5 — ChrisGPT · 2026-10-27
- ChatGPT keeps telling tech-support users to email 404 Media's Jason Koebler — petrusenko_max · 2026-10-27
- GPT-6.1 Terra called the awkward middle tier, priced at half of GPT-6.1 Sol — cedric_chee · 2026-10-27
- Mystery model "Terra" scores 48 on Artificial Analysis index with no public API — haider1 · 2026-10-27
- GPT-6.1 Terra (Max) Appears on Artificial Analysis Scoring 48 — theanonymousone · 2026-10-27