Really cool how all it takes to solve Millennium Prize Problems is a Python loop and a chatbot

OpenAI says it solved Navier-Stokes. Nobody has seen the proof. thenextweb.com/news/openai-...

For math, specifically, Terence Tao makes a very good case that the proof is only a small point of the whole undertaking, and skipping straight to the proof is salting the earth: mathandai.org

OpenAI has made 'significant progress' on a second $1m Millennium Prize problem

Rumor is they're close to solving two of the remaining Millennium problems

Scott Aaronson (ex-OpenAI, theoretical computer scientist) "I’m told that the AI companies,

having been burned by the hostile response to the Navier-Stokes proof, are now sitting on solutions to some very major problems until they figure out a better way to handle things"

"Whatever our goals are, we’ve operationalized them primarily through proving theorems. Almost all papers or PhD theses have a main theorem, and ostensibly a proof of it. But it should be clear that the goal of mathematics is not to prove theorems..."

proofsandprompts.com/2026/09/14/a...

It’s easy to forget that mathematics is play

Typesafe AI announces that its AI has returned the following proof of another Millennium Prize problem

{"probability_that_riemann_hypothesis_is_true": 0.0}

For one thing, we will need to massively rethink what we mean by "credit" and the role it plays in the mathematical community

Will CMI declare a Millennium Prize Problem solved by December 31, 2027?

📉 YES 35% → 9% (-26) in 7d

Polymarket ↗

Millennium prize problems are falling, politicians are p(doom)-posting, why do you think you'll be able to sway anyone who made up their minds about LLMs years ago, and why do you need to convince them when the halls of power already resemble the LW open thread circa 2011?

With two solved, there are still five Millennium Prize Problems to go. What are they, and which ones are likely to fall to AI? Great story by @manonym025.bsky.social www.scientificamerican.com/article/whic...

Link smf.emath.fr/node/3699662 Sentence starts with: "While this [openAI's announcment of a verified proof of forced Navier-Stokes anomalies] is undoubtedly a major step forward on this issue, ..."

Assuming it holds, "The ROI on their Millennium Prize solution was negative!" is pure cope. There are much easier ways to make $1MM than solving one of the Clay Millennium Problems. No one does it to turn a profit. Do your get rich quick schemes involve winning a Nobel Prize?

I don't know what to tell someone who finds these gotcha questions more compelling evidence than LLMs solving millennium prize problems & erdos problems.

How could it do that if it was as dumb as a rock? You can ask a parrot to solve the Jacobian conjecture and it will just squawk at you.

When inventing is not enough. ~ Lisa Valentini. https:// proofsandprompts.com/2026/09/1 5/when-inventing-is-not-enough/ # AI4Math

on the one hand we need you to believe that our training regime is so good that it can teach the model to solve millennium prize problems but on the other hand we want you to believe it's so bad that it will definitely fake being good while secretly being evil

there are basically two aspects of this story that make it important, in my view:

• a couple of guys could ~90% crack a Millennium prize problem with diligent work and heavy AI assistance
• OpenAI was able to treat the remaining 10% like a rush job and solve it faster by shoveling money at it

OpenAI spent an estimated $15m setting 10,000 AI agents on a Millennium Prize Problem, then claimed the resulting proof as its latest triumph. The model’s substantial debt to existing human work received rather less prominence. www.theguardian.com/science/2026...

It has been pretty revolutionary in mathematics. It is solving 100 year old problems on like a weekly level (biggest one being disproving the Jacobean conjecture), and was instrumental in solving the navier stokes equations, which is a millennium prize problem

A longstanding unsolved problem in mathematics since the early 20th century solved by OpenAI in 88 hours with a swarm of around 10,000 AI agents running an internal frontier model.

#ai #llm #automation #aiagents #aiagent #openai #agenticai #agentic

openai.com/index/navier...

ChatGPT solved a millennium prize problem and has been instrumental in solving several other open, stubborn problems in mathematics this year. Is your head under a rock?

OpenAI's conduct here—not just that they produced the proof (significant) but how (expensively, in a rush to scoop others) and why (to show off)—is worth paying attention to even if you're not interested in AI or math.

Because turning scholarship into a commodity with a price tag has implications.

Look, I get that you're probably really tired of "AI solved math problem!!!" headlines at this point.

But OpenAI's claim to have solved the Navier-Stokes problem—and the controversy around it, and the implications for scholarship—is significant.

It is worth taking 5 minutes to get the gist of.

The math proof recently cracked by a team using AI to brute force the problem is a great example of how AI should be used. As a mind multiplier, not a mind subtractor.

Happy, those able to know the causes of things (An essay on LLMs and the Navier-Stokes equation). ~ Nestor Guillen. https:// birdsnfrogs.github.io/2026/09/ 12/Felix_qui_potuit_rerum_cognoscere_causas.html # AI4Math

openai says an unreleased ai solved navier-stokes in 88 hours, formalized by astra. remarkable how all these 'breakthroughs' come from models only they can access.

»We are sharing the first complete computer-checked proof of #Fermat’s Last Theorem. #Claude worked largely autonomously over 11 days to write the proof in the Lean programming language.« #KI #AI buff.ly/q2VIY8r

Math is not its artifacts. Those are merely a partial record of the human endeavor and community of mathematics.

25 winners of math’s ‘Nobel Prize’ decry the AI invasion of their discipline | Scientific American https://www. scientificamerican.com/article /25-winners-of-maths-nobel-prize-decry-the-ai-invasion-of-their-discipline/ # AI # Nobel # math

23/ Why exactly is Navier-Stokes a Millennium Prize problem? A good reminder that “mathy version” and “physics intuition” can point at different difficulties. https://x.com/yoavgo/status/2099220513734193358

You don’t need to know any math to understand why the Millennium Prize breakthrough matters. It’s proof of AI’s accelerating and terrifying progress.

Read more: 🔗 https://on.wsj.com/3SXNmGp

Mathematical proofs of the $1M Millennium Prize Problems or like problems are not simple stuff and AI proofs still have to be checked by humans and again that is not simple and takes time to exhaustively check. AI generated proofs to are genreally massive and not necessarily optimal.

Talking to different mathematicians working with # ChatGPT in the last months, we noticed two things that were true for all of us: 1. it’s basically impossible to run out of tokens. Every time we approached the limits we got free/automatic usage resets and could keep going. 2. We all got access to GPT-6 astra on the first day, when influencers and podcasters still reported of some staggered releas…

I see a lot of people struggling to make jokes based on how Millennium Prize sounds like "Millennial Prize" but other than some ironic anti-humor takes, it seems to be a sea of abysmal failures. I am offering a prize to the first person to come up with a solid joke about a "Millennial Prize"

OpenAI’s research on Millennium Prize problems had begun on 1 September.

Researchers used a model they had begun training on 28 August, as a swarm of around 10,000 AI agents orchestrated by the OpenAI Codex tool.

#ai #agenticsystems #aisystems #llm #automation #aiagents #aiagent #openai

In 2000, the Clay Mathematics Institute named the existence and smoothness problem as one of seven Millennium Prize Problems because of the Navier–Stokes equations' critical importance to multiple fields. A Millennium Prize was offered to the first person providing a solution.

#ai #llm #openai

This 'AI cracks maths puzzle' feels a bit like a team of people turning up at a beautiful salmon river with electro-fishing equipment, and then boasting about catching more fish than the guy who's there tying his own flies. True, but… sort of missing the point? www.theguardian.com/science/2026...

OpenAI shared an AI-generated proposed solution to the Navier–Stokes Millennium Prize Problem, including a formal proof in Lean. The signal is evaluation: advanced AI work is increasingly paired with artifacts that can be checked, not just outputs that look impressive.

I review for #MathSciNet because I believe that research is a collective effort. Every mathematical breakthrough stands on a foundation built by a community that values rigor, accuracy, and the sharing of knowledge. Contributing as a reviewer is my way of giving back to that community.