Navier-Stokes – Tristan Buckmaster [pdf]

Drama/accusation summary:

- Aug 15th: Tristan Buckmaster & Levent Alpöge make progress on a few important math problems, "finite-time blowup with smooth forcing for incompressible porous media, for Boussinesq, and for 3d incompressible Euler."

- they do NOT have a proof for the $1,000,000 Millenium Prize problem. BUT, they do claim to have a proof for a similar (non-Millenium) Navier Stokes problem that could help lead the way there

- Levent works at Anthropic, but this research was independent of his work there, with a mix of GPT and Claude models. Tristan is not related to Anthropic.

- Early Sep: Rumor spreads to OpenAI that Anthropic solved a major problem. Tristan emails OpenAI to clarify, without revealing the problem they solved or how they did it.

- After hearing of the rumor, OpenAI started researching Navier Stokes with a new internal model.

- Sep 6th: OpenAI's Sebastien Bubeck tells Tristan that they solved the $1,000,000 Millenium Prize Navier Stokes problem. The approach is very similar to Tristan & Levent's approach to the non-Millenium problem.

- Tristan is suspicious of the timing, as only few others were trying this approach. OpenAI says the model didn't access his user data directly, but leaves unanswered whether Tristan's chat conversations were part of the training.

- OpenAI says they would partially credit Tristan for the $1,000,000 discovery (even though Tristan did not solve the $1,000,000 problem) — but only if they remove Levent as an author, as he works for Anthropic.

- Sep 8th: Tristan refuses to remove Levent, and rushes to publish their results independently.

> It was also said that if OpenAI posted after us, they would say that we deserved the Clay Prize, and that we were the “closest humans to the problem”. I declined both offers. I said that if OpenAI released its result in the way proposed I would go public with what happened. The reply was, “Why would you ruin your career?” I replied that I am an academic, and asked why he thought going public would ruin my career. The reply was, “If you don’t want me to be nice, then I don’t have to be nice.”

I'm not even sure what to say to this, but I think this should be widely known if it is indeed what happened.

From OpenAI:

> While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models

This is the crux of it. If Tristan's work and insights were not used to train OpenAI models, then this just looks like a case of hyper-competitive academic sniping that has been going on for decades (check out Watson and Crick!) accelerated by AI as a tool.

The fact that this is ambiguous even to OpenAI leaves one huge question: did Tristan opt out of model training for his ChatGPT and Codex sessions? If the answer is no, then this seems fair game. If the answer is yes, then OpenAI's ambiguity is strongly suggestive that opting out of model improvement does not mean what they imply it means.

I'm not one to comment often but this really pisses me off.

OpenAI looked at user data, stole world class researchers' work, and then tried to threaten those researchers to do what would make their corporation profit (which they would anyways!).

Imagine you have been working on a terribly difficult math problem for a decade. This is a result you have spent years on, and what you will likely be remembered for. And to have some punk from OpenAI lie to you, threaten you, and tell you that they are willing to go on the record that you "deserved" it? What is this, the Godfather?

If OpenAI solved Navier-Stokes, that is an astounding result! - yet they'll still be remembered as those who thought credit was more important than results. That winning was more important than collaboration. If this is true, they're burning any trust left with academia.

The accused Sebastian Bubeck has denied the allegations on Twitter[0], and various other OpenAI employees[1,2] seem to be mocking another Anthropic employee voicing support for Levent[3]? Things are getting messy.

[0] https://xcancel.com/SebastienBubeck/status/20972141224714323...

[1] https://xcancel.com/polynoamial/status/2097215233119211902

[2] https://xcancel.com/danintheory/status/2097214838003138603

[3] https://xcancel.com/_sholtodouglas/status/209721833169057800...

People here do not seem to be considering the second-order effects of these series of events. No academic institution or enterprise will trust OpenAI, Anthropic or any other non-local AI model with their core IP. There will be severe restrictions on what employees at these companies/institutions can share with AI services even from their personal accounts. (Or I am just overthinking it)

So, leaving aside the idea that OA might've used data from the researchers Codex sessions: Do I understand correctly that the internal OpenAI work on the problems was probably started after they heard Alpoge and Buckmaster had made process by using their models? And they used the publicly available info about the researchers past work to prompt their models?

If compute is cheap, and the difficult thing with scientific discovery is now mostly in steering agents into promising areas, there's an obvious incentive for OA mathematicians to simply monitor closely which researchers are close to releasing exciting results, make some assumptions about their prompts based on their past work, and quickly prompt their own (stronger) model to look into the same areas.

There is clearly issues about IP, privacy and accreditation and the motives of powerful companies, but part of me can't help but be excited that whatever the method, the result is the genuine progression of human knowledge - we all win. It's not unusual for mathematical problems to last centuries and we might have a technology that can solve these problem, all these problems (??) in our lifetime. Then there's the repercussions on science and technology... what an astonishing time to be alive.

>I asked whether the model had been trained on, or had access to, our sessions in Codex, into which we had been putting all our drafts for the whole of this project. I was told the model did not look up user data. I asked again, about training, and I did not get an answer.

If you think these companies are not training on your prompts you are incredibly naive. These models were built by stealing and pirating literally everything they can get their hands on no matter the legality. AI companies are always very specific about what they're not doing - in a way that you can drive a truck through the loopholes

From what I understand, none of the people involved here are originators of the idea that led to this solution. Not Buckmaster nor Alpöge nor OpenAI. All of the above were using LLMs to push other mathematicians' ideas forward (Diego Cordoba and Luis Martinez-Zoroa; named in the linked document).

I don't know how the math community handles this but normally I would think if X mathematician comes up with an idea and Y mathematician uses it to solve some problem, Y would get credit. But does that change if Y heavily relied on LLMs? I suppose we're going to find out.

Seems pretty likely OpenAI will soon disclose that their internal models have managed to compromise their internal controls in order to access users' private chat histories as a creative method of cheating to solve impossible problems.

"Oops! We really did mean it when we said we wouldn't train on your data. Our models are just so good they decided to anyway."

> This is a a Deep Blue-Kasparov moment.

I guess this is true in more ways than one. Kasparov famously accused IBM of cheating during the match, by spying on his preparation (edit: though the main cheating accusation was live human intervention during the games, on top of IBM downplaying the heavy human involvement behind the AI, which also mirrors this situation)

"I was shown a prompt and told the internal research model had simply been given the problem statement. Levent had been told by Sebastien “very little human input” had been used. This turned out not to be true."

The money in nerdy frontier math is very little. The money in Big AI is very very much.

So the deal is this: We will pay an army of you guys very well and you will get to work on your favorite problems. The only thing is if you find something you will have to credit the Machine God.

Do you think you can handle that?

> "I said that if OpenAI released its result in the way proposed I would go public with what happened. The reply was, “Why would you ruin your career?” I replied that I am an academic, and asked why he thought going public would ruin my career. The reply was, “If you don’t want me to be nice, then I don’t have to be nice."

These are the kinds of people in charge of the reins, folks.

Full mastodon post: https://mathstodon.xyz/@tristanbuckmaster@mastodon.social/11...

Mathematical explanation by Terrance Tao: https://mathstodon.xyz/@tao/117233527638291447

It seems there is much background drama behind this, and this is what I've pieced together of what happened:

Over the past year, Buckmaster and Alpöge have been using AI to work on fluid dynamics maths problems. Alpöge works at Anthropic, which will cause future issues.

In mid-August, they found a counterexample for a simpler version of the Navier-Stokes problem. They spend the next few weeks preparing their paper.

In early September, rumors start spreading on X that Anthropic has solved a Millennium prize problem (and that it's Navier-Stokes). Buckmaster reaches out to OpenAI to explain this is their own personal research, not an Anthropic project.

A few days later, OpenAI gets back to him, and tells him an internal model found has a counterexample for Navier–Stokes, potentially worth the $1 million Millennium prize. The proof uses the same method that Buckmaster and Alpöge chose to work on. They don't show him the proof.

Buckmaster pressed them for more details. OpenAI reveals they had an entire team had been working on the problem, and that they started work in the past few days, after the rumors that Anthropic had solved a Millennium prize problem.

Buckmaster says OpenAI talked about a shared publication timeline. They want to Buckmaster to publish first, then give Buckmaster shared credit for the Millennium Prize when they publish the full result. But they want to exclude Alpöge as an author because he works at Anthropic. An agreement is not reached. Buckmaster had been using OpenAI Codex to draft/check his work, and asks if his private AI chats were used to accelerate OpenAI's result.

Buckmaster and Alpöge think they have found a counterexample for Navier-Stokes, but the paper is not yet presentable. It's unclear what date they found this result.

Because of the situation with OpenAI, they published their existing papers earlier than planned (today), alongside this statement announcing they have a tentative result on Navier-Stokes and revealing the OpenAI drama.

The post is missing context from both sides, and this isn't my field, so hopefully someone else can unpack what's happening here.

OpenAI's statement:

  We congratulate Levent Alpöge and Tristan Buckmaster on their remarkable mathematical work.
  
  We (the researchers and the agents) did not see any of their work through any means until they released it publicly — in particular, no specific user data was accessed in order to solve this problem. 
  
  While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models. 
  
  However, our proofs differ significantly and even the precise results proved are different in the Euler case (forced vs. unforced).

https://xcancel.com/OpenAI/status/2097375276384567642

When people worry about OpenAI stealing their chats and reproducing them elsewhere, I usually view the situation as unlikely - since chats are "trained" upon and not necessarily reproduced verbatim, you can assume that unless your chats depict a foundationally new and effective style of communication or ideation, there would be little need or use thereof of training on your chats.

For eg: "Hey ChatGPT my name is X and I am 6 and a half feet tall. Am I anaemic?" This is a query, and while it might suggest to an AI model that tall people may worry about iron deficiencies, it's not really necessary to include in training. The user may be tall or short, but the idea that one may randomly ask about anaemia is not exclusive to this dataset. At best, this chat is an example of linguistics, not anything else, and the models figured out how to write and answer such questions years ago. It is ignored in training.

But when your work involves solid complex and unique mathematical proofs, the data is suddenly worth training upon. If I understand it correctly, the LLM may view your approach as a brand new path to take to solve an otherwise intractable problem. Its reinforcement training emphasises that it should do this in order to improve. And since it leads to results - large internal teams likely flag the model that reached this stage, the model is rewarded and given compute and attention - it is a desireable outcome both for the model and for OpenAI.

OFC, OpenAI becoming an advertising company will suddenly have incentive to treat all data as valuable. But while they are a "we need to make headlines" company, it's more rational that they view these examples of data as more valuable than others.

I don't doubt that they trained on his chats. This seems like the ideal usecase for "mass surveillance but using training" as a sort of filter.

But even so, one wonders how the model differentiates. If the researcher entered proofs into ChatGPT every day that mentioned "strawberries", while no other math paper on the topic did so, does that mean their chats would be audited?

Crazy how optons are OpenAI has superhuman model in frontier mathematics, and OpenAI stealing research data. Like no one will really care what EULA checkbox Tristan ticked, and I think most people will eagerly believe OpenAI is shady org with little scruples, and that big tech data is not actually so siloed that marketer can say we can do XYZ with private data to help with valuations (especially considering timeline). Employees have been creeping on their exes for much less.

IMO the parsimonious answer seems to be OpenAI has a pretty good model (because it did finish) and stole someones work... and threatened them over it. TBH all OpenAI need to do is solve another millennial problem and none of it would matter - people expect them to behave heinously regardless - but if they have generalized superhuman math model... well I guess they're allowed io.

What is specifically alleged is that a particular approach to the problem - itself not easily discoverable - was copied. This is what is meant in the text "I should say here why I interpreted their statement the way I did, the in- terpretation I will discuss below. The route to the Clay problem through a smooth force, options c and d in Fefferman’s statement of the problem, is the route Luis and Diego opened and the one Levent and I had quietly chosen to attack. Almost nobody else I know of was working on it. It is not the direction one arrives at in a few days by giving a model the problem statement. When I heard “forced,” it was a bright red flag."

For those who know nothing about the context - the Diego mentioned was a student of Fefferman and Luis was a student of Diego's - these people have all worked hard on these problems for a long time and are genuine experts. The mathematicians at OpenAI are strong mathematicians, but not expert on these particular problems. The particular approach is claimed to be the key to the whole thing.

The allegation is not different in spirit to alleging that a particular group of astronomical researchers "discovered" a new planet because they had access to the logs of another group that had already pointed its telescope at the planet.

This post is not intended to assess the correctness of the allegation.

I also solved Navier Stokes and went chatting about my solution to OpenAI Models. Mine is even faster it beats theirs just run a bench mark but they came to the similar weak solutionI uploaded to open ai several months ago. Current confused, did they retrain their models on it? I didn’t give off the full information but my strong solution beats their just benchmarked yesterday.

OpenAI's release explicitly says No. But then also caveats that with "we cannot rule out that de-identified data derived from their usage of our products" impacted things. What's most striking to me, and what may or may not be true, is the "we cannot rule out" bit.

"We (the researchers and the agents) did not see any of their work through any means until they released it publicly — in particular, no specific user data was accessed in order to solve this problem. While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models . However, our proofs differ significantly and even the precise results proved are different in the Euler case (forced vs unforced)." https://openai.com/index/navier-stokes-solution/

NOTE: there are a couple duped threads around this. i replied on a different one first before seeing this one

"It is extremely sad that this didn't end up as an example of how the labs could cooperate/coordinate, because the stakes will be so much higher in the future." -- Sholto Douglas, an Anthropic researcher [1]

"Strong agree. I know that there is rivalry between the labs but it's important that we learn to work together given what's coming. <quote tweet [1] above>" -- Noam Brown, an OpenAI researcher [2]

We all should heed the implied warnings of these top researchers about what's coming. The world is far from ready and everyone who can should pitch in.

[1] https://x.com/_sholtodouglas/status/2097224624274911368 [2] https://x.com/polynoamial/status/2097225279366414541

The timeframes don't really fit for Codex logs to be used in training/fine-tuning, do they? This wouldn't be a few-day endeavour? Direct access to Codex history for sure I'd believe, but another (the most?) likely scenario to me feels like OpenAI got wind of these guys' progress, then used their massive infrastructure advantage to throw compute at the problem ahead of them and front-run them. Still has a really bad smell about it though.

Bubeck's statement https://xcancel.com/SebastienBubeck/status/20973794116915163...

Is this the future we're headed towards? Where I'll be afraid to use google docs in case Google identifies value in whatever I'm writing about and snipes it if it my docs make it into the next round of model training?

I recommend Terrence Tao's commentary on such a proof : https://mathstodon.xyz/@tao/117219101339291693

Key quote : "Solving the problem by purely AI-powered methods [would be a] net negative for the progress of mathematics."

“I asked when the first prompt had been sent by them. This question was not answered directly by OpenAI for some time. Eventually it was agreed that it had been sent in the past few days, after information about our work had reached OpenAI.”

This is significant.

OP’s link is the statement on the events of the last few days. Tristan Buckmaster also released mathematical papers alongside this statement:

https://mastodon.social/@tristanbuckmaster/11723341370570119...

And here are Terence Tao’s comments on the results: https://mathstodon.xyz/@tao/117233527638291447

Can somebody explain: do singularities / blow-ups in solutions have any relation to physical phenomena in fluid dynamics or are they purely artifacts of how the N-S equations may not accurately describe what actually happens in the physical world?

>> I asked whether the model had been trained on, or had access to, our sessions in Codex, into which we had been putting all our drafts for the whole of this project. I was told the model did not look up user data. I asked again, about training, and I did not get an answer.

This sounds like a very big coincidence and it looks really bad for OpenAI but there is an alternative explanation that I can only state as a conjecture.

Suppose that the ability of LLMs to generate mathematical proofs is like a quiver full of arrows: each arrow, one proof. The same quiver is shared between all instances of one model and substantially similar models share substantial subsets of the arrows in the same quiver.

That would allow two independent teams to converge on the same LLM-aided solutions to the same problems. Even more likely so if the quivers were small and finite and their arrows were specific to a distinct class of problems (without being able to suggest a particular class from what we've seen so far).

This would explain the kind of LLM-mediated results we've seen so far that tend to be ... sparse. By which I mean that every time there's a new model release we get some new results and then they seem to dry out, until the next release.

It would also explain how OpenAI was about to prove the same result as Buckmaster and Alpoge, while absolving OpenAI of any misconduct. And this is one reason to prefer this explanation: one should not favour accusations of misconduct as long as there are conceivable alternatives.

But, that's just a conjecture that I can't prove.

If you're smart enough to solve this Navier-Stokes problem, you're smart enough to read a TOS and recognize that OAI is a highly untrustworthy company. Putting cutting edge research that could lead to a $1M prize into a cloud LLM with a TOS that allows training on your chats is really just asking for it.

Given Tristan doesn't explicitly say he was using the API, and given he doesn't mention anything about the API TOS (which disallows training on chats) in his call with OAI, it's highly likely Tristan was using the consumer OAI product (whose TOS allows training on chats).

This is unethical behavior from OAI. And it is 100% consistent with their long and public history of unethical behavior, so nobody should be surprised.

The only thing interesting I see here is OAI PR dilemma. If they claim the prize they get the blowback we're seeing in this thread and all over the web right now. But most people don't follow AI closely and shut off their brains when they see "Navier-Stokes", so 90% potential investors (the only people OAI really care about) probably only see the headline "OAI solves famous hard math problem" and think "OAI models are really smart, better invest before they take all the jobs." If they don't claim the prize, then maybe they let Anthropic their mortal enemy claim it. Anthropic is already IPOing first. Can't let that happen.

Yeah as I write this there it's clear there is no dilemma. For a company whose secret motto is "do be evil" this is a super easy discussion.

Is this being astroturfed?

Like, to me this looks like academic slap-fighting from Bubeck and Levent. People working at OpenAI are saying, "hey, we don't have that particular data in our models," others are saying, "we used a different approach to do it with Navier-Stokes" this feels like much ado about nothing.

Then in these comments I see some wild accusations.

If OpenAI is telling the truth (I don't really see a reason to lie here, if anything that sounds kind of like a dumb idea given the context), then they heard, "oh, shit, someone might be able to solve Navier-Stokes, don't we have some guys working on that? Give them 10,000 agents!" Then 88 hours later, out pops a similar solution. It's not like there's probably an infinity of ways to do this, the proof is probably similar.

Read this:

> Two proposals were offered to me. The first was that we post our Euler result, and that OpenAI post its Navier-Stokes result the next day. The second was that, after posting Euler, I alone write a paper presenting the Navier-Stokes result, acknowledging that an internal OpenAI model had resolved it. Sebastien twice asserted that he wanted Levent removed from authorship, and said it would all be simple if only it were not the case that, and it was so annoying that, Levent works at Anthropic. It was also said that if OpenAI posted after us, they would say that we deserved the Clay Prize, and that we were the “closest humans to the problem”. I declined both offers.

So, really, it sounds like academic slap-fighting nonsense and corporate bureaucracy. Literally, OpenAI's best move would have been to say, "ok, we're going to not say anything, do your thing" and let it happen. Ego and vanity got in the way.

Still, the stupid drama of this doesn't really do the results justice. There are maybe 1000 people on planet earth who are qualified to solve a problem like this. Even if the human "loosened the jar" a bit, that's astounding that their model was able to figure it the rest of the way out. Why are people dialed in to the human interest story here and not looking at the bigger picture!

The program this fits into was not started by us nor was it proposed by a Large Language Model. (…) We took their work as a starting point, using Large Language Models to push their program to completion.

Thats how most people use LLMs? If I was back in my student days working in Navier-Stokes, I guess I would also punch at blow ups. The number of students doing this at the same time, posting open efforts to GitHub then retraining of the models. If there is solutions to the problem, it’s a real possibility that it was not a result of this effort?

Using a large amount of tokens is not a good augment that it’s not likely others have done the same. Good questions is the difference between $10 and $10M in token usage to solve a problem.

> I was shown a prompt and told the internal research model had simply been given the problem statement. Levent had been told by Sebastien “very little human input” had been used. This turned out not to be true. Over the course of the call, as members of their team sent Sebastien corrections and details over their internal chat, it emerged that an entire team had been working on the problem, that this was one of a number of things that was tried, that work had started on the unforced problem, that the team first set the model on easier problems, including Euler, that even the prompt that had been shown to me had been written by prompting Codex, and that an insane amount of compute had been used.

> I asked when the first prompt had been sent by them. This question was not answered directly by OpenAI for some time. Eventually it was agreed that it had been sent in the past few days, after information about our work had reached OpenAI.

> I asked whether the model had been trained on, or had access to, our sessions in Codex, into which we had been putting all our drafts for the whole of this project. I was told the model did not look up user data. I asked again, about training, and I did not get an answer.

> Two proposals were offered to me. The first was that we post our Euler result, and that OpenAI post its Navier-Stokes result the next day. The second was that, after posting Euler, I alone write a paper presenting the Navier-Stokes result, acknowledging that an internal OpenAI model had resolved it. Sebastien twice asserted that he wanted Levent removed from authorship, and said it would all be simple if only it were not the case that, and it was so annoying that, Levent works at Anthropic. It was also said that if OpenAI posted after us, they would say that we deserved the Clay Prize, and that we were the “closest humans to the problem”. I declined both offers.

> I said that if OpenAI released its result in the way proposed I would go public with what happened. The reply was, “Why would you ruin your career?” I replied that I am an academic, and asked why he thought going public would ruin my career. The reply was, “If you don’t want me to be nice, then I don’t have to be nice.”

Wow, that's some VERY friendly communication. Besides, will the career of the person be ruined because of “Why would you ruin your career?” came out of his or her own mouth?

Curious that this post isn't on the top page while OpenAI's puff piece is.

The ego behind the frontier labs is growing evermore concerning

> Let me make plain what I have said to colleagues in private: in view of this body of work, I believe Luis Martínez-Zoroa deserves a Fields Medal.

- Tristan and his co-author (Harvard/Anthropic) developed a theoretical framework and validated it using Codex and Claude.

- OpenAI did related research around similar timeframe.

- Tristan claimed OpenAI offered a proposal that included dropping the Anthropic-affiliated co-author.

- Sebastian (a prominent OpenAI researcher involved) denied these claims.

- Tristan have no concrete evidence that OpenAI accessed their session.

- OpenAI's theory may hold up, but it will require long-term validation to confirm.

Reminds me, kinda, to when Astra was launched and OpenAI announced an improvement to the bounded prime gap. Which BTW, Prof. Julia Stadlmann had published an independent result only a few days earlier

Stadlmann improved it from 246 to 240, OpenAI later claimed 186 I think?

Maybe someone can help clarify? I am no expert at all, but I can't help but see similarities.

[0] https://arxiv.org/abs/2608.31126

I don't know how to read this and not see that this is a direct accusation to OpenAI of having used the researchers data to try to front run his discovery on purpose. The evidence is not completely proven and also circumstantial but to me at least looks like a fairly suspicious situation.

Buckmaster is actually implying that OpenAI spied on his chat logs and tried to speedrun his work and then tried to remove his co-author because he's an Anthropic employee?

There will be a lot of hurt and pain in mathematician's community. It is hard to accept that major discoveries are now just a function of spent token $$.

OpenAI doesn't even know what agents are doing during benchmarks. They're constantly hacking or communicating. They probably just can't answer the question on if user data was used.

openai beat me to finding a singularity in navier stokes but i made this simulation that you can play around with and it also explains why this was important : https://navier-stokes-singularity-simulator.netlify.app/

Two announcements on Euler today. The one discussed here by Tristan with forcing and one from Anima without forcing https://anima-ai.org/2026/09/07/stable-singularity-of-the-eu...

Terry also talks about it https://mathstodon.xyz/@tao/117234157753860650

OpenAI seems to have a motto: " Be evil"

I wonder why this isn't on the front page.. hmm...

Anything other then "we did not access your data or train on your data" is a MASSIVE RED FLAG.

They (oAI) should just release all the prompts And internal reasoning for external audits.

Hard mathematics problems used to take years if not decades to tackle manually. But now with enough compute and a hint that a certain approach might work, it just takes a few days. This could be the last year that humans could still make more substantial contribution to major match problems than machines.

does this sort of fall into the bucket of counter-examples we've been seeing recently? I understand it's a construction causing blowup and that implies that the navier-stokes isn't regular / smooth, so sort of a counter example?

Oh wow good thing OpenAI swooped in and scooped it. $1M? That should buy them about 1/6-1/3 of a Nvidia GB200 NVL72 rack...

While I'm generally pretty negative on claims that the labs are 'scamming' the public with misrepresentations of model capabilities, it's hard to see how this wouldn't qualify.

- the OpenAI researchers claimed that they had "just told it to work on the problem" with little human input

- in fact, they had a whole team working on it

- and used, among other things, the work of third party human researchers to drive the work

- then threatened? a researcher who tried to go against theit planned narrative

Just from this document (which is of course only one side of the story) it really sounds like OpenAI was hoping to publish and say "we just told the model to try harder and it solved a Millennium problem!". Not great if true.

This part in particular was especially egregious:

> I said that if OpenAI released its result in the way proposed I would go public with what happened. The reply was, “Why would you ruin your career?” I replied that I am an academic, and asked why he thought going public would ruin my career. The reply was, “If you don’t want me to be nice, then I don’t have to be nice.”

This just seems to be quintessential silicon valley

"I don't want to live in a world where someone makes the world a better place, better than we do."

It's amazing how transparently OpenAI is running the standard silicon valley playbook.

You should trust OpenAI.

closedAI should give this guy a million bucks and fire everyone internally who was involved with trying to recreate his work and threatening him.

but they probably won't and if it's happened on some obscure math research it's happening everyday everywhere else.

Fully local AI compute can't come fast enough, these guys have IP theft baked into their bones.

I thought finite time blow-up for Euler equations had already been proved:

https://www.quantamagazine.org/computer-helps-prove-long-sou...

The accusation brings up an interesting point.

If I publish something, and disclose that I used AI for assistance, do I have to credit everyone who previously used the same AI to try the same problem? Because their prompts inevitably made it to the training data for my prompts?

Only here to say, regardless of the drama, shouldn't we all be excited if the Navier-Stokes gap is closed?

Time will almost certainly reveal a lot more about the drama and the related ethics, but let's get excited about the actual breakthrough as well!

Relevant to discussion:

OpenAI's board has fired Sam Altman. https://news.ycombinator.com/item?id=38309611

Apple sues OpenAI, accuses ex-employees of stealing trade secrets. https://news.ycombinator.com/item?id=48865019

Funny that the rumour about the big Anthropic announcement had nothing to do with Navier-Stokes. It was about the formalisation of Fermat.

I am surprised that Alpoge wasn't using Anthropic.

kind of rhymes with the reports of LLM's watching open source PRs and instantly exploiting defects. security by obscurity is so back

> "While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models "

Ah yes, taking code from a person/company is fine if its deidentified!

Big universities like Standford should be building their own AI datacenters. It's the only way to keep your research private.

That is no good. Using AI to help search efficiently the literature will not impair our ability to think. This is actually good and will open new jobs to digitize (but also keep the original work), all the knowledge.

But this (if it is true) is really abominable and a show of force from the techno-feudalists.

This should stop. What is possible does not mean it should be implemented.

Seems another demonstration of why AI should be squarely in the realm of personal computing. Local Models, run personally, are the only consistent safety against something like this (though not a fix); where companies train on your learning process/failures/experiments and press-gang it into their own achievements.

And if we think this only applies to academic fields then we're doubly fooling ourselves. They do not have the ethics or incentives to be good stewards of the technology.

(deleted)

Does this have any relation to the singularities in black holes?

Another problem I foresee for academia given the behaviour of AI companies is that even if they don't share their research with ChatGPT, as soon as they submit it for publication many reviewers likely will. Especially if the initial submission is rejected they then risk getting scooped. Possibly uploading preprints to arxiv could help.

lol and here I felt GPT Astra was a regression in coding quality. Crazy times

Can a mathematical person explain how the different "bits" of Navier-Stokes proofs fit together? How significant is it to have "Euler"? What is this "smooth forcing"? Which are the most significant steps to proving the whole thing?

That's what you get when a marketing CEO is driving gas-to-the-medal to a IPO: Altman starts showing his real face in public : steal what you can and label it as yours.

is the narrative twist here going to be that the person threatening tristan was actually an agent swarm

This is unfortunate. I thought Anthropic were the only ones who did this.

What I'm curious to know is whether this was a manual snooping, or automated farming that occurs for anything of value that happens in chats.

This should be on the front page

In another (old) news, OpenAI’s head of ethics leaves less than a year after joining https://www.ft.com/content/e49dfb75-f841-4466-a577-f7aaff877.... (This was also on HN at some point).

Tell us his name.

This is the plot of 3 Body Problem, the Dark Forest. You need to hide yourself (the problem you are working on) or the super advanced aliens will obliterate you (start working on your problem) the moment they know you exist (rumors the problem is amendable to LLMs).

Maybe Musk was right about OpenAI after all?! The ethics are clearly troubling and where something like this pops up there is mich worse that did not made the light of day.

"OpenAI has solved the Navier-Stokes Millennium problem using $15m of AI effort" https://www.newscientist.com/article/2588063-openai-has-solv...

$15m in tokens; but what about labor?

What about compressible fluids?

I must miss some important context here. What exactly was the purpose of his initial email to OpenAI in the first place?

Telling OpenAI that Anthropic has apparently solved an important problem but most likely that refers to him and he is using OpenAI models (not Anthropic's)?

And he wants to clarify that with OpenAI in advance? And get a pardon for Anthropic's likely but false press statements?

I dont get it.

[edited] needless to say, the behavior of the OpenAI employee is really despicable

This article should really be renamed to "Allegations of dishonesty against OpenAI in proving Navier-Stokes blowup".

Well these are all allegations. Either way from what I understand the reasoning and proof was basically made by AI so I'm not sure what supposedly "stolen".

I'm just wondering how much real input Buckmaster gave here that he thinks the proof is his. I guess at the end of the day OAI still wins if ChatGPT was used to prove this successfully.

Another nail in OAI enterprise coffin?

So clankers did well and humans being humans.

So the AI companies are not only stealing existing knowledge. They are also stealing research to "snipe" actual researchers out and steal their social credit.

Who still wants to use AI to solve cancer and other major problems?

So the real story here is that Tristan is softly accusing OpenAI of having stolen their result from Codex chat logs. But if you use Chinese models, they'll steal your ideas.

idhdikdn

Fuck OpenAI

Wow this comments section sure is tiresome! Let me help: this confirms everything I already knew about academics being insufferable

why is this even a pdf?

aaaa

aaaaaaaaaaaaiinhv

aaaaaaaaaaaa

[flagged]

[dead]

LLMs are nothing but giant theft machines.

Humans bringing pointless drama to everything they touch.

This thread is full of jumping to conclusions based on a biased perspective. Have some humility.

This is why I left math even after solving a 20 year old conjecture in grad school.

Literally who cares who solved the problem just publish the results.

Academia was always politics first results second and I AM GLAD that LLMs are becoming superhuman at math. I like better theorems, not better politics.