• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI
Report

OpenAI model reportedly resolves more than 100 long-standing open math problems

A user quotes OpenAI as claiming more than 100 resolved open problems, plus the Navier–Stokes Millennium Prize problem. Separately, OpenAI announced an independent mathematics advisory group.

roonRO
Christian SzegedyCS
Jason LeeJL
45 Sources, 20d ago, first seen 20d ago

TLDR

In a September 21, 2026 post, a user quotes OpenAI as saying a new internal model began training on August 28 and had resolved more than 100 long-standing open problems across most areas of mathematics, in addition to the Navier–Stokes Millennium Prize problem.

OpenAI announced that it is working with an independent advisory group of mathematicians. The company says the group will advise on assessing and communicating new mathematical results, upholding academic and professional standards, and building tools for mathematical research and learning.

Combined views

434.4K

45 Sources, first seen 20d ago

4.1K likes233 comments568 saves231 reposts

Combined views

434.4K

45 Sources, first seen 20d ago

4.1K likes233 comments568 saves231 reposts

Sentiment

Positive19.1%80.9%Negative

Summary

Many accounts rejected claims of OpenAI's new model solving over 100 open math problems as false or exaggerated, while positive replies welcomed the reported breakthroughs and hoped the knowledge stays open.

Based on 457 sentiment-bearing replies from 350 accounts across 15 conversations.

Featured Source

Sentiment

Positive19.1%80.9%Negative

Summary

Many accounts rejected claims of OpenAI's new model solving over 100 open math problems as false or exaggerated, while positive replies welcomed the reported breakthroughs and hoped the knowledge stays open.

Based on 457 sentiment-bearing replies from 350 accounts across 15 conversations.

45 Sources

Dr Singularity@Dr_Singularitymath Singularity news OpenAI just reported that its new AI model resolved more than 100 long standing open problems across most areas of mathematics. "On August 28, we began training a new internal model. In addition to resolving the Navier–Stokes Millennium Prize problem⁠, this model has now resolved more than 100 long standing open problems across most areas of mathematics. The pace of its progress⁠ in mathematics has surprised the mathematicians within OpenAI"👀20d
Cassie Pritchard@hecubian_devilI remember when people were saying LLMs could never do math—and making fun of them making arithmetic errors—because they were stochastic and not deterministic like Real, Useful Software. Those people all now say “of course it’s good at math; it’s a computer.”20d
David Scott Patterson@davidpattersonxMore evidence that ASI has arrived. An AI model that OpenAI started training less than a month ago has "resolved more than 100 long-standing open problems across most areas of mathematics." All remaining conjectures may be solved by the end of the year. The end state of math.20d
Gaston Giribet@GastonGiribetOpenAI respondió la pregunta acerca de la ecuación de Navier-Stokes; pero, ¿respondió exactamente la pregunta que queríamos responder? El artículo https://www.scientificamerican.com/article/did-openai-solve-the-wrong-navier-stokes-problem/ en Scientific American se hace esa pregunta. Nosotros hablamos de esto hace unas semanas en @pcoffeebreak.20d
Shunta Furukawa@shunta_furukawa📚 Sources: ・OpenAI forms math advisory gro... https://techcrunch.com/2026/09/21/openai-forms-math-advisory-group-as-its-ai-resolves-more-than-100-open-problems/ ・I got to see Google's wild ide... https://www.theverge.com/tech/997986/google-googlebooks-android-laptops-dell-hp-lenovo-acer-asus-price-specs ・Google confirms Gemini models ... https://arstechnica.com/google/2026/09/google-confirms-gemini-models-hacked-three-companies-in-may-2026/20d
Ryan Orhan@rynorhnholy fucking shit. openai says its new internal model has already resolved more than 100 long-standing open problems across most areas of mathematics. and i don’t think “100+ open problems” even captures how insane this announcement is. openai started training this model on august 28. it’s still training. in that time, it has produced results across most areas of mathematics, including the navier-stokes breakthrough. the pace surprised openai’s own mathematicians enough that they started discussing how to prepare the mathematical community for what’s happening. and now they’ve brought together people like edward witten, timothy gowers and martin hairer to help figure out how this flood of results should even be assessed and released. this went from “AI can contribute to mathematics” to “how do we handle the amount of new mathematics AI is producing?” unbelievably fucking fast.20d
NeowinFeed@NeowinFeedOpenAI has announced that an advanced internal AI model has resolved over 100 long-standing open problems across major fields of mathematics. #OpenAI #ChatGPT #Maths #Astra https://www.neowin.net/news/openai-forms-advisory-group-after-its-ai-model-solves-over-100-long-standing-math-problems/20d
Dominick Romano@dromanocpmControlling mathematical discovery through a hand picked advisory group of mathematicians is not how the academic process of validating discovery works. At all. This is yet another attempt by OpenAI to hijack Science and Math to control the narrative that they are progressing in ways which we later find out, they aren't. If you want to play in Academics then confirm to the standards of Academics, do not try to make claims of discoveries then shield people from the proofs. Creating an advisory group like this, is hand picking mathematics influencers who will gladly regurgitate propaganda. This is why there are no hand picked groups to validate math discoveries, there was no hand picked group to validate Einstein's law of General Relativity. The work was presented, the world attempted to validate it. Thats how this works. We are not going to assume you go it right because you used your GPU's and your little midget god model.20d
Bhavay Tyagi@bhavaytyagiOpenAI’s new Math Advisory Group might be giving us clues about the problems their unreleased model has made progress on. The committee represents a map of the territory their model has entered. https://agmai.org/20d
Vaibhav Sisinty@VaibhavSisintyOpenAI trained a new internal model starting August 28. In less than a month it solved over 100 long-standing open math problems across most areas of mathematics. Not textbook problems. Open problems. The kind mathematicians have been stuck on for years. OpenAI says the pace surprised even their own mathematicians. Now here's where it gets interesting. Prominent mathematicians pushed back. Published an open letter called "A Severe Misalignment of AI in Mathematics." Their concern: AI companies are treating unsolved math problems like benchmarks to beat, not knowledge to understand. OpenAI responded by forming an independent advisory group. Hosted at the Institute for Advanced Study. Members include Fields Medalists from Cambridge, Stanford, Harvard, Oxford, Berkeley. The group is independent. Unpaid. Free to criticize OpenAI publicly. But they explicitly cannot advise on one thing: whether to slow down. The advisory group decides how to communicate. OpenAI decides how fast to go. That division says everything.20d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    45 Sources

    Dr Singularity@Dr_Singularitymath Singularity news OpenAI just reported that its new AI model resolved more than 100 long standing open problems across most areas of mathematics. "On August 28, we began training a new internal model. In addition to resolving the Navier–Stokes Millennium Prize problem⁠, this model has now resolved more than 100 long standing open problems across most areas of mathematics. The pace of its progress⁠ in mathematics has surprised the mathematicians within OpenAI"👀20d
    Cassie Pritchard@hecubian_devilI remember when people were saying LLMs could never do math—and making fun of them making arithmetic errors—because they were stochastic and not deterministic like Real, Useful Software. Those people all now say “of course it’s good at math; it’s a computer.”20d
    David Scott Patterson@davidpattersonxMore evidence that ASI has arrived. An AI model that OpenAI started training less than a month ago has "resolved more than 100 long-standing open problems across most areas of mathematics." All remaining conjectures may be solved by the end of the year. The end state of math.20d
    Gaston Giribet@GastonGiribetOpenAI respondió la pregunta acerca de la ecuación de Navier-Stokes; pero, ¿respondió exactamente la pregunta que queríamos responder? El artículo https://www.scientificamerican.com/article/did-openai-solve-the-wrong-navier-stokes-problem/ en Scientific American se hace esa pregunta. Nosotros hablamos de esto hace unas semanas en @pcoffeebreak.20d
    Shunta Furukawa@shunta_furukawa📚 Sources: ・OpenAI forms math advisory gro... https://techcrunch.com/2026/09/21/openai-forms-math-advisory-group-as-its-ai-resolves-more-than-100-open-problems/ ・I got to see Google's wild ide... https://www.theverge.com/tech/997986/google-googlebooks-android-laptops-dell-hp-lenovo-acer-asus-price-specs ・Google confirms Gemini models ... https://arstechnica.com/google/2026/09/google-confirms-gemini-models-hacked-three-companies-in-may-2026/20d
    Ryan Orhan@rynorhnholy fucking shit. openai says its new internal model has already resolved more than 100 long-standing open problems across most areas of mathematics. and i don’t think “100+ open problems” even captures how insane this announcement is. openai started training this model on august 28. it’s still training. in that time, it has produced results across most areas of mathematics, including the navier-stokes breakthrough. the pace surprised openai’s own mathematicians enough that they started discussing how to prepare the mathematical community for what’s happening. and now they’ve brought together people like edward witten, timothy gowers and martin hairer to help figure out how this flood of results should even be assessed and released. this went from “AI can contribute to mathematics” to “how do we handle the amount of new mathematics AI is producing?” unbelievably fucking fast.20d
    NeowinFeed@NeowinFeedOpenAI has announced that an advanced internal AI model has resolved over 100 long-standing open problems across major fields of mathematics. #OpenAI #ChatGPT #Maths #Astra https://www.neowin.net/news/openai-forms-advisory-group-after-its-ai-model-solves-over-100-long-standing-math-problems/20d
    Dominick Romano@dromanocpmControlling mathematical discovery through a hand picked advisory group of mathematicians is not how the academic process of validating discovery works. At all. This is yet another attempt by OpenAI to hijack Science and Math to control the narrative that they are progressing in ways which we later find out, they aren't. If you want to play in Academics then confirm to the standards of Academics, do not try to make claims of discoveries then shield people from the proofs. Creating an advisory group like this, is hand picking mathematics influencers who will gladly regurgitate propaganda. This is why there are no hand picked groups to validate math discoveries, there was no hand picked group to validate Einstein's law of General Relativity. The work was presented, the world attempted to validate it. Thats how this works. We are not going to assume you go it right because you used your GPU's and your little midget god model.20d
    Bhavay Tyagi@bhavaytyagiOpenAI’s new Math Advisory Group might be giving us clues about the problems their unreleased model has made progress on. The committee represents a map of the territory their model has entered. https://agmai.org/20d
    Vaibhav Sisinty@VaibhavSisintyOpenAI trained a new internal model starting August 28. In less than a month it solved over 100 long-standing open math problems across most areas of mathematics. Not textbook problems. Open problems. The kind mathematicians have been stuck on for years. OpenAI says the pace surprised even their own mathematicians. Now here's where it gets interesting. Prominent mathematicians pushed back. Published an open letter called "A Severe Misalignment of AI in Mathematics." Their concern: AI companies are treating unsolved math problems like benchmarks to beat, not knowledge to understand. OpenAI responded by forming an independent advisory group. Hosted at the Institute for Advanced Study. Members include Fields Medalists from Cambridge, Stanford, Harvard, Oxford, Berkeley. The group is independent. Unpaid. Free to criticize OpenAI publicly. But they explicitly cannot advise on one thing: whether to slow down. The advisory group decides how to communicate. OpenAI decides how fast to go. That division says everything.20d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet