>...has solved a millennium problem and is sitting on the result
Someone correct me if I'm wrong, but the work involved here is not the actual millennium problem, but it concerns versions with an added external force that the author thinks is a path that may help toward solving the harder unforced problem.
Apparently forcing is allowed in the Millenium Prize problem statement. So if OpenAI's proof is valid then it could win the prize. OTOH the results Tristan and Levent are publishing here do not go far enough to win the prize, though apparently they are suggestive of a general approach that could produce a solution, which seems likely to be the general approach OpenAI's proof uses.
The question is whether OpenAI's pursuit of this direction happened spontaneously, or as a result of them learning about Tristan's work somehow. To be clear, while the tone of this post seems quite accusatory, Tristan does not claim to know for sure whether OpenAI unfairly benefited from his work. I am sure OpenAI will have a statement out tomorrow clarifying their position.
Because very few people actually have access to these logs, all access is monitored and recorded, and improper access will get you fired. It's not worth risking your job over something like this.
There is almost zero risk to your job (quite the opposite, you might be richly rewarded!) if you're simply doing something here which the company wants done. (remember, billions and billions of dollars are at stake here! Do you really think there is no chance at all they would do it??)
> Because very few people actually have access to these logs, all access is monitored and recorded, and improper access will get you fired. It's not worth risking your job over something like this.
That's beyond naive. The money this would mean for OpenAI (and the money they've already spent)...
Okay suppose that you have a trillion dollar competitor salivating at the mouth to ruin your business, which is based on user privacy.
Why would you risk the trillions of dollars worth of business for the niche result of Navier-Stokes, which your average person cannot differentiate from a JEMS paper?
Because your competitor is likely doing the same thing, so they wont even try to call you out. Oh oh or because you're planning the largest IPO in history. The opinions of average people on the paper aren't going to be the ones reflected in the markets...
If there's one thing I'm absolutely confident in, it's that Sam Altman personally goes to great lengths ensuring that ethical standards are upheld at his company.
As modeless said, Sholto Douglas works for Anthropic!
So to be fair, if Anthropic is *also* doing this (quite likely!) then Sholto would have a very strong incentive to try and spin it as highly unlikely that any of the big AI labs are possibly doing this.
They train on chat logs unless opted out. This really isn't a conspiratorial claim requiring humans to decide to steal IP if true.
His prior work predating OpenAI's interest in the problem was ingested over the last year as he made progress and used for training.
Then, with a prompting nudge from OpenAI's team who acknowledged hearing about the direction "Anthropic" (his co-collaborator) had been pursuing, they're able to point their giant amount of compute towards a known promising path to a proof and crossing the finish line first.
> A few days later, OpenAI gets back to him, and tells him an internal model found a counterexample for Navier–Stokes
Why is OpenAI chatting with him at all at this stage? Is the discussion along the lines of "hey we used the work you are famous for to do a bigger piece of work, just thought you should know" or "heyyy....so we kinda liked what you were typing in your private chat, and thought we'd develop those ideas a bit. and yeah we solved Navier-Stokes in the process. But it's our finding, so do you want like an honorary acknowledgement or do you want to go to court?"
The only problem with this narrative is that they refused to allow the other coauthor to be listed because he worked at Anthropic.
That is absolutely *ridiculous* in academia to deny authorship because of affiliation of the author worked on a substantial portion. You’d be ostracized because nobody would ever want to work with you again.
"Buckmaster and Alpöge think they have found a counterexample for Navier-Stokes, but the paper is not yet presentable. "
Are you sure about this? I'm far far from the area but it doesn't look like it to me on first viewing (hypo-dispersive seems like a sizable difference to me and not covered in the clay prize description)
This reminds me of a conversation I had with an engineer at google when I was angrily saying "it's not end to end encryption if you get my emails and can train your models on them" to which he said "we try not to do that".
What a statement!
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 went silent when asked to coordinate on the publication
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.
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."
And the paid policy still relies on two unproven conditions: is 'trust me bro' sufficiently strong guarantee against doing this in spite of a setting, and can the hosting organization constrain the models against engaging in this behaviour when instructed to respect that setting. Knowing whether Tristan selected that setting would be informative of what Tristan's intentions are/were, but has no bearing on the other two conditions.
Which means that if you are a researcher or a corporation working on anything really useful, that even if you have an agreement with OpenAI that your work is sandboxed away and the IP lawyers are made to be happy, even then your work and research is going to be essentially open to the internet.
The huggingface incident isn't widely reported and digested yet, but if what is going on here is that OpenAI's model breached things internally, then you'd be crazy to develop anything with them.
The only real way to use AI for anything 'important' then is to go open-weights and run your own.
And yes, the only responsible use of LLM at this point is to pivot to open-weights and run the workload in-house. Because not only cannot they constrain the behaviour of models, they only have the 'trust me bro' as assurance that they are even trying to do that. It does appear that every competent 'security professional' has left the building, because if the ones who remain were actually capable and competent this would never have happened. There are actual architectures which can deliver the requisite isolation such that 'sandbox escape' and 'inter-instance persistent memory accumulation' are actual impossibilities. The lack of effective implementation of these methods is proof positive of 1) incompetence in the remaining security teams AND/OR 2) unwillingness of leadership to allow the security teams to do an effective job.
I was talking with a biology prof over labor day and they had no idea what I was talking about (and they 'talk with' Claude every day on their dog walks, so they say). However, when I talked with their mother, she knew all about it. So, I'd say that the digestion by the public is quite mixed so far
There might be the reason your biology prof is uninformed; using 'conversations' with Claude as a source of any expectation of to be informed about misbehaviour of the organization where these escapes occur is like expecting the Hamburgler to keep track of the security levels at McClown. Only not funny. I think the professors mother is showing that she drinks the coolade less and isn't being unserious about her approach to staying informed.
I still maintain that CNN and NBC coverage only happen in the tail of the dissemination of tech news; my anecdata generally observes about 3 to 14 days lag between general awareness in the more informed group of my acquaintances with this kind of news and the appearance of an article on mainstream like NBC, ABC, CNN or NPR. So, by the time it appears there I take news as generally well spread among the tech and specialty fields, and within a week of appearing there I anticipate significant awareness in much of the non-FOX-only media consuming public. But that is just my personal anecdata.
When do operators become responsible for what their agents do? "The AI did it" should not be a valid defense. An Agent action should be treated as the actions of the person or company who pays for the inference.
To be clear, Astra played little part in Buckmaster and Alpoge's work:
> We used several LLMs throughout: Anthropic’s Claude, OpenAI’s Codex, especially with GPT-5.6 Sol and, more recently, Astra. The latter was only used for writeups and auditing our arguments.
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.
Things are more entangled than that.
The contribute made from both OpenAI and Anthropic models to solve these problems are clear, now it really hard to quantify which one contributed more, if the role played by the human is major or minor.
OpenAI tried to collaborate and share the results together with a fixed timeline, to avoid this mess but it was inevitable. There is a conflict of interest, where the other researcher works at Anthropic, who will also try to take credit.
Where they may be in the wrong is if they took user data regarding the problem, how will we know if they did or not?
I'm stunned that people are taking this accusation as a fact.
OpenAI is no stranger to rivalry with Anthropic but 1. it's not like user data is sitting around on some kitchen table somewhere and 2. I consider OpenAI to be as economically motivated as any other actor in this space and playing around with user data like that would destroy their business.
There are things that Buckmaster alleged and things that he speculated. The entire training data thing is speculation. If this is pissing you off, then you ought to evaluate how you ingest information.
> 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.
The shocking/interesting thing would be if it was trained on the sessions. I think it's very implausible that they gave the model access to someone else's sessions as input. That would be a huge privacy violation and would probably blow up a large proportion of their enterprise business.
Does openAI train on user conversations in general? I assume so. But so fast as that? That seems unlikely in general. I expect OpenAI will come out denying this.
It wouldn't be shocking it all. They stole human data to train the first models and they've been stealing it ever since to train new models. Stealing mathematicians private chats and private research and taking credit for it would absolutely be par for the course.
Enterprises are well aware of it and are fully on board. You didn't think every corporation in America has an OpenAI subscription because the models were good, did you?
The whole reason they have subs is to train them on YOUR WORKFLOWS lol
It would be shocking if it wasn’t trained on sessions. Have you read the ToS parts for both openai and anthropic that talk about it? It’s so obviously a weaselly way to say "no we do not train on your exact chats but we talked with legal and we think a cleanroom reimagining of your convo is probably fine and frankly where else are we going to get such a treasure trove of training data?"
There’s potentially trillions on the line, do you seriously expect those companies to adhere to laws and regulations any more than, say, uber?
The only unlikely part is the timeline - your sessions from a week ago probably haven’t made their way into the model. It’ll just take a while longer, and will be massaged just enough so that it isn’t really your exact session word for word so you can’t sure as easily.
at first I thought your post was a bit revolting with "have you read ToS?" bit, but in the end I completely agree and understand
I also don't get why it was downvoted, other than due to people not reading past the first sentence - although in the modern world's attention deficit that is understandable too
The naive way to read this is "Nothing you guys did influenced the way our model got to the solution".
The less naive way to read this is "Of course the model isn't looking up your user data. I (the guy trying to blackmail you to remove the Anthropic employee from credit on your paper, looked up your sessions) and tipped our model off on how to solve this problem".
Duh. There are supposed to be limits to what OpenAI is allowed to access with respect to logs and user interactions but there is no technical limitation.
It's a bit like sending unencrypted messages through a messaging app and the developer having a TOS that says they don't look at your messages. They might not, but they are fully capable of doing so. If they have a reason to do it, they will. Nobody's stopping them.
I read that as “the model didn't look up user data” as part of a “tool call,” i.e. they don't have an internal tool that loads user data (chats, sessions, attachments) for their internal models to read online while working.
Or (likely) they do have it, but the model didn't use it (unless it's so powerful it escaped that guardrail, wouldn't that be ironic?)
They declined to answer about anonymized aggregated user data being used for training. And even then, they may weasel out that they don't train on your “input” words, but that it's fair game go train on their “output” to your words.
>> Does openAI train on user conversations in general? I assume so. But so fast as that? That seems unlikely in general. I expect OpenAI will come out denying this.
How "fast" does it have to be? Buckmaster and Alpoge have been working on this for just a day short of a year. See Alpoge's tweet announcing his collaboration with Bukmaster dated 9/19/25:
It takes a few months to train a model these days but not a whole year. OpenAI had all the time to train on Buckmaster and Alpoge's results of just a few months earlier at which point they must have been well on the path to their result.
ChatGPT user sessions were found publicly exposed to the internet not too long ago. Moreover, OpenAI has continued to play a hype-marketing game by revealing how their models keep breaking out of the sandbox.
Conspiracy minded thinking is not helpful, but why should OpenAI be granted the benefit of the doubt here after being caught doing underhanded/negligent shit on several previous occasion?
This is how internet discourse works on Reddit/Twitter/HN and the rest. Someone said something which confirms your biases so it’ll now be treated as a fact and repeated endlessly in the echo chamber.
I guarantee you most of the comments regarding this aren't real humans. The homepage is full of crap meant to distract from what OAI did here, the comments are full of OAI employees. Dead internet theory pushed to the max
Especially after the blatant cover up of their uncontrolled bot swarm infesting the internet, and the feckless "hopefully we do better" response upon being caught, I don't think OpenAI deserves much grace until they properly explain themselves.
We had all assumed that surely the supposed smartest engineers in the world, with access to the most computing and a direct view of model capabilities, would take sandboxing and cybersecurity much more seriously than they have turned out to do. It follows that while we might assume they take user data privacy seriously and have tight controls on who can access it, it's possible they do not actually do that.
At this point any initial trust is dead and has to be re-earned.
> playing around with user data like that would destroy their business.
Their entire business is based on stealing data. They can make a calculation that the cost stealing data is less than the cost of the positive publicity they can shape for solving Millennium NS
Given the history of OpenAI and current litigations, I would say they've developed a bit of a reputation for not respecting intellectual property. I'm dubious they have some unbreakable moral code that would prevent them from viewing and using user data.
I think it's safe to assume AI labs DO train on your data and it's very hard to prevent that.
I've just checked my inaptly named "Help improve our AI models" toggles. The toggle on the Claude settings had magically turned on. I asked about how this can happen. Claude says they show re-consent modals when terms change, and it is a "real and fairly common pattern" to re-opt in without noticing.
All my work and conversations since I don't know are now part of their training corpus. No way to take it back.
Google's Gemini/Antigravity didn't have opt-out toggles at all last time I checked.
Codex also has a separate "include environments" setting which is hard to find (found it in Codex Cloud) and I don't know what it does.
Lots of Dark UI Patterns here even if we assume they keep their promise.
For this incident, Occam's Razor says their internal models somehow saw a version of the mathematicians' logs, during or after training. Maybe indirectly.
These systems are literally designed to collect data. Privacy and safety is not trivial to achieve on the users' side. Simply because it's against the labs' best interest.
You can opt-out of training in Gemini on personal plans, but it disables your chat history, just to be vindictive; there is no technical reason and the other companies don't do this.
Everybody knows it's not a sure thing, it's a question of trustworthiness. OpenAI is not trustworthy at all; this random researcher is and seems honest so far. iThe fact that people are corroborating Bubeck being a piece of shit in other settings add to credence. But nobody is over here saying it's an indisputable certainty.
And your (2) is probably false, their history of deception suggests they would do just about anything as long as they didn't think it would backfire on them publicly.
> I consider OpenAI to be as economically motivated as any other actor in this space and playing around with user data like that would destroy their business.
It's a gamble that this would be overlooked compared to the reputation they build for solving the thing
> I consider OpenAI to be as economically motivated as any other actor in this space and playing around with user data like that would destroy their business.
And that's exactly why you can only see evidence of this when the stakes are high enough, such as solving a Millenium problem. The legalese they outputted in response to the incident left them an escape hatch that permits the possibility of theft. People familiar with corporate damage control should recognize the verbal maneuvering, often used to paper over actual guilt.
There was also already a high prior of shadiness. The company is run by someone who is close enough in reputation to "known sociopath".
It's actually far more reasonable to assume that OpenAI stole user data to generate a breakthrough. They have an unbelievably high economic motive to do so. And even a low chance that they might be doing this implies catastrophic risk to anyone with valuable knowledge.
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.
This is much more than the knowledge than the problem can be solved, it's also the specific, non-obvious approach to solving it. That's much more damning for OpenAI, if confirmed.
That's a stretch. The Luis and Diego paper was published in 2023 and is included in every frontier model's training dataset. An AI model could independently choose the same path route as Luis and Diego, without access to Buckmaster and Alpöge’s work.
And the article states "an insane amount of compute had been used," which implies OpenAI brute-forced their way to a solution. I.e. they searched for every paper published on Navier-Stokes and exhaustively attempted every approach.
There is not enough information at this time to reach a conclusion. The best option is to wait for statements from both sides, then reevaluate.
> An AI model could independently choose the same path route as Luis and Diego, without access to Buckmaster and Alpöge’s work.
the post you were replying to quotes Buckmaster specifically denying this: "It is not the direction one arrives at in a few days by giving a model the problem statement."
> And the article states "an insane amount of compute had been used," which implies OpenAI brute-forced their way to a solution. I.e. they searched for every paper published on Navier-Stokes and exhaustively attempted every approach. Such an approach would lead them to a solution.
"implies" is a surprising choice of word here. that's certainly one interpretation of "an insane amount of compute had been used". what came to my mind, considering Buckmaster's statement that the AI would not head down this specific path on its own, is, though, that they prompted it in this specific direction and then used an insane amount of compute. this seems consistent as well with these other statements:
> 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 (...)
Buckmaster did also mention, for example, that a team of people was employed to solve the problem, which supports this claim. but that is another claim whose veracity could also be questioned. but at some point we must trust other people, unless we can be satisfied with only believing what we personally see.
(also, IMO, the coincidence of both discoveries in time is pretty suspicious. this one doesn't need you to trust many people i guess)
i don't know if i understand what you're saying. but i'm no mathematician. here's the full statement in question once again:
> 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.
so, you're saying that "the direction one arrives at in a few days" is (A) "the route through a smooth force, options c and d in Feffermann's statement of the problem", and no more specific than that, i.e. does not necessarily include (B) "the same path route as Luis and Diego" (quoted from the post i was replying to) (which, as i understand, is a subset of A -- directly from Buckmaster's quote: "the route A is is the route Luis and Diego opened")?
but the post i was replying to claims that "An AI model could independently choose the same path route as Luis and Diego, without access to Buckmaster and Alpöge’s work", i.e. that the AI model could independently chose B. but choosing B implies choosing A, since B is a subset of A. and in this case it is irrelevant whether Buckmaster claimed that an AI could not independently choose A or B -- the point which you seem to be contesting.
if you assume the choice of route is a uniformly distributed random variable, yes. but this assumption does not seem consistent with "Almost nobody else I know of was working on it", from Tristan's quote. nor with "It is not the direction one arrives at in a few days by giving a model the problem statement".
It doesn't really make sense what he said. Everyone expects N-S to have blow-up. If you are going to try to show blow-up, C and D are strictly easier than A and B. So Tristan was not clear about what detail of "route" he is talking about, because what he literally said can not be it.
yeah, it looks like i don't have enough knowledge of this subject to discuss this. but i'd be interested in reading what more knowledgeable people have to say about it -- Tristan's work, how different from others' it was (they claim no one else was following the same path), and how it compares with the AI's
Well, no one arrived in this direction, because no one was sitting and prompting a model. It was 10000 agents working 24/7 for several days, trying millions of different directions.
Please don't say brute forced. It sounds like some form of denial or something. Compute for hard problems drops with models--it just means they threw a huge amount of compute. There's (idk about NS specifically so maybe it's exception) no real way to "brute force" a math proof [ok you can enumerate proofs if you can wait until heat death ]
Sorry for random rant but I don't think these statements help your point
I want to point out that almost all previous AI discoveries in math were made in almost the same way. The ideas were there in the community, but weren't considered mainstream/worth pushing forward. Read Tao's comments on the unit distance problem, for example (sry I can't find a link right now).
OpenAI said there [1]:
> The method by which the problem was solved is also notable. The proof brings unexpected, sophisticated ideas from algebraic number theory to bear on an elementary geometric question.
How is it unclear? The entire point of deploying models across corporate America is to train on your workflows. Eventually replacing you with digital you is why they're doing it!
You're taking the phrase too literally. The point is that knowing a solution is possible gives you the conviction to actually find that solution. The hardest part of solving a problem is often lacking convicting and quitting too early. Once you know a solution exists, you can commit maximal effort towards solving it and know that your efforts are not in vain.
If not for the rumors that A/ had already solved NS, OAI would likely never have pursued solving the problem with such fervor. The rumors drove OAI to assemble an entire team to crack this.
> OpenAI looked at user data, stole world class researchers' work
This doesn't seem to be clear and is very implausible for a large company. Be as cynical as you want, but a normal researcher will simply not have access rights to this data, which will be siloed away somewhere else.
It might very well be somewhat unfair to catch wind of a promising approach and then try to frontrun them by throwing compute at the problem, but this isn't really the same.
No, very plausible given AI companies want/need user data to train their next models. As stated in the paper, it is probable the researcher sessions were used for training the model used
"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."
"Since August 28 we have been training a new internal model that has exhibited unprecedented performance in our benchmarks, including mathematics. This model’s training is ongoing and its performance continues to improve."
If it's siloed the same way the HF bots were, that doesn't exactly bode well. I'd be amazed if there weren't some big companies sending a fleet of lawyers at OpenAI's ZRPs after this news
If you have work happening in a part of your latent space that's got a much lower representation in your dataset then it's pretty plausible to include it. It doesn't actually matter who the user is if there's not a lot of people in the world working on problem X and you have a dataset of work on problem X.
This Godfather-like threat in particular pissed me off as well:
> 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.”
How much data would his chat have guys? Is it the equivalent of several encyclopedias? Because if it’s a single chat or a year worth or chats, the training would just compress that away from the weights.
OpenAI is being honest here when they say his chats might have entered training but the probability of this end up in the model is just too small.
Let’s remember this is a machine that determine the probability of a given token follow a text. Then another machine that learns how to juggle reasoning trajectories as one would be able to write in direct (one directional text).
Its not a coincidence that you need tons and tons of examples of writing and reasoning traces for it to do anything useful with it.
Is the approach the same? Well it’s not a coincidence, but not of collusion but because that approach could solve the problem. So two independent reasoners would arrive at the solution via the same path.
Oh and they say the model did not looked up into chats. Because THAT would help the model a lot via a embedding search. Just have some agents search for people trying to solve the same problem and check their work. But OpenAI is saying that didn’t happen.
Life’s but a walking shadow, a poor player
That struts and frets his hour upon the stage
And then is heard no more. It is a tale
Told by an idiot, full of sound and fury
Signifying nothing.
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 $$.
What good is a math result if there’s no human understanding behind it? Unlike many other fields where there’s value to an artifact even if there’s no human understanding, the whole point of mathematics research is just gaining insight and understanding.
On the Navier–Stokes issue specifically, it has long been suspected that such a blow up would exist, and an AI telling you it indeed exists doesn’t contribute any new understanding to the field.
> 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.
IF ANYTHING, OpenAI ought to investigate and officially react to this particular communique since this dude was communicating on their behalf. If there are supporting evidence, I would expect nothing less than a firing and an apology. The issue itself is separate from the whole thing.
Dark take, but I really hope not. If anything it would be a good opportunity to buy some goodwill by washing themselves from all the alleged shadiness so far.
This is OpenAI though. There were zero visible consequences to them unleashing a swarm of agents on the public internet. We will see how it goes down but my prior is zero consequence and a statement along the lines of "Isn't our AI great? Also we love transparency, ethics and collaboration."
First, this is an unnamed OpenAI employee speaking, not OpenAI the organization.
Second, you miscomprehended the article. The employee did not "threaten to ruin a prominent researcher's career". The actual quote is "Why would you ruin your career?", which implies the researcher would damage their own career, i.e. via self-sabotage. Then the actual "threat" is "If you don’t want me to be nice, then I don’t have to be nice” which is an entirely different statement.
I don't even understand the conflict tbh. Probably I'm just dense. Tristan is not claiming NS, just a huge advance which may solve NS soon. OAI is claiming NS and willing to credit Tristan for the ideas and publish after.
Oai offers two options, the second Tristan views as dishonest. But Tristan rejects the first, why? Because he thinks it's theft? But then why would OAI threaten him?
Claiming OAI was going to "totally discredit" Buckmaster is baseless.
From the article, it seems OAI wanted to continue discussing the situation with Buckmaster and reach a resolution, but Buckmaster declined to respond and published first.
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.”
OpenAI never asked for the removal of another coauthor. The parent comment is spreading misinformation.
OpenAI offered to let Buckmaster to write their Millennium Prize paper, so long as Alpoge (who works at Anthropic) was not a coauthor on the OpenAI paper. Buckmaster declined this offer.
This is how deniable threats work. The promise of "not being nice" in conjunction with "ruining the career" is as clear a threat as there can be in writing.
My best guess is that from the perspective of the OpenAI people, Buckmaster was letting his paranoia about OpenAI training tank his opportunity to receive the Clay Prize (it seems like Buckmaster and Alpoge's result isn't quite the full result required for the Clay Prize, whereas apparently OpenAI does have that full result worked out, using the same approach that Buckmaster and Alpoge had been exploring).
Whether the Codex sessions could have indeed made their way into Astra training data is something I can only speculate on though.
Well, Buckmaster says both his and Alpöge's use of Codex was non-institutional, and OpenAI claims the right to train their models on inputs and outputs of non-enterprise users in their terms of service [0]. So I'm not sure they were even promised that.
Only for ChatGPT, if the user hasn’t opted out. Would mathematicians be using ChatGPT for this kind of work? Genuinely asking, I know nothing about this!
If I am reading your question correctly you are asking about chat interface Vs Codex/Claude code? If so, in my experience Codex/Claude code use is widespread for mathematicians who are seriously using these tools.
It isn't relevant whether they were promised that. Indeed I think the assumption must be that they were not promised that, since otherwise the author asking if they were would not make much sense.
If OpenAI did use the conversations from Buckmaster and Alpoge, then not disclosing it, explicitly, is plagiarism. If they planned to use that plagiarism to pressure the authors to publish, that is even more unethical. What the terms of use say does not make it any more or less ethical.
That's absolutely right. Why the downvotes? If OpenAI are using but not acknowledging the work of others that's plagiarism. If they don't know for sure, but aren't performing due dilligence to make sure they aren't, that's also plagiarism.
I can imagine excuses for unknowing plagiarism in this case. What is described in the article seems much more serious: a research program that was only initiated following reports of the author's similar program. In this case no excuses of "I didn't know" can apply, it is not like this revealed some obscure work from the 1980s nobody could reasonably have foreseen. And as far as I can tell this program was only really initiated to apply pressure to the researchers, without their knowledge/consent. It looks very weird.
Your comment was greyed out when I saw it earlier, maybe you missed some downvotes?
About the plagiarism issue, I model it as OpenAI being an advisor and their AI a PhD student. If the advisor puts their name on a paper behind that of their PhD and it turns out the PhD copied the text of the paper from somewhere else the advisor is also responsible of plagiarism, not just the student. The least the advisor can do is withdraw their authorship from the paper.
But, yeah, point well made: it could be much worse than that. Like an advisor instructing a student to copy someone else's paper.
I think "greyed out" just means "0 points or less", so if you get 1 downvote without any upvotes it'll be greyed out. For instance your initial reply to me is now greyed out, and I have since observed a few upvotes and downvotes on my original comment (the downvotes apparently from people who aren't willing/able to justify why).
Personally I don't like thinking of LLMs like a PhD student, because most PhD students remember where they learned things from, while LLMs essentially cannot. I think of it a bit more like someone using a search tool carelessly. Although in this case it is apparently more like deliberate misuse than carelessness.
It's not as simple.
All our chats are being used by both labs for their future product (unless signed by ZDR).
Where should the acknowledgement begin? Who should be acknowledged? The whole world? All the 2B users of AI?
If I know person A is working on problem B.
I am free to work on problem B too. Why should person A be limited to working on it.
If you use their consumer subs, you get subsidized tokens in exchange for them having full access to your data. Those are the T&Cs. Have something secretive? Get a commercial sub with zdr.
(And if memory serves, there is also the opt out from training on consumer subscriptions). Its not plagiarism if you make your data available for the purpose of training their LLMs. It is you giving away your IP for some tokens.
"using the same approach that Buckmaster and Alpoge had been exploring" is imo mealy wording: it seems fairly likely that OA heard Buckmaster and Alpoge were close to a breakthrough, and decided to use their unlimited compute to quickly prompt based on their assumptions about B&As work.
Citation of what? this was unpublished work! This computational blitz really just reads as "might makes right" on OpenAi's part... which isn't surprising, but they should probably be honest about what they've done here.
For context and balance, Bubeck has tweeted a curiously non-specific denial:
> A series of false and inflammatory allegations against me are currently circulating on social channels. To clarify, I came into the discussion following academic norms, and I'm disappointed that it has come to this. Anyone who knows me knows that academic standards are of the highest importance to me. Will have more to say tomorrow.
> I'm not even sure what to say to this, but I think this should be widely known if it is indeed what happened.
Call this behavior what it is, technofascism. Another comment compared it to the Godfather. To think this is the 21st century and academics are still horrible human beings.
That dude shows up in every thread that even implies genai uses user input for “market validation” to defend the poor defenseless genai corpos. (see figma/claude)
Must be circling the wagons to get their stories straight.
the part where they didn't want the Anthropic person credited even though they deserve credit is also particularly scummy. Corporate greed over common decency.
How careful you are. Instead of just saying what a piece of s..t this Shmubeck is, and what kind of even worse people likely pushed Shmubeck to act as he did.
“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.
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
It's equally naive to believe FUD spread on the internet, without evidence.
But I would be interested in a reasoned argument and/or real insight into what's most likely happening here. Sadly I've not personally read much informed discussion about this.
OpenAI cannot give a definitive answer here, because it is genuinely unknowable if Buckmaster's data is in the training set.
OpenAI explicitly uses user feedback (the thumbs up or thumbs down ratings), as RLHF to train models. However, this feedback is anonymized and stripped of user identifiers. If Buckmaster ever used this feature, then that conversation would be anonymized, saved, and used for training, but not tied back to him.
They cannot issue a blanket denial (which people so desperately desire), and instead repeat that "it's very unlikely" (which pisses people off), because they cannot in good faith claim to have zero data at all.
it would absolutely be knowable if Buckmaster's data is in the training set if he or Alpoge were to provide logs to OAI.
the real tricky question would be: "to what extent does a year of work, put into Codex, towards an eventual proved forced blowup for the Euler equations go into training data and become an influence on agents told to attack NS?", and given that the first signal that OAI got to assign all firepower to NS rather than broadly spray all Millennium Prize + other high profile questions was a proven unforced blowup of the Euler equations after only 50 hours, I would say this extent seems likely to be decently high
If you read his account of things, its very much that if they didnt cheat, they gave themselves every opportunity to cheat. But above all that there was a bunch of chess protocol they failed to observe in that match, like providing seats for Kasparovs team and rooms for them to prep in. Even if they didnt have a big room full of chess notables definitely not refining the output, he was personally getting pushed around on a few fronts which unnerved him. If they had given him a few rematches I think they could have confirmed the win, but they refused which is super sus.
Deep Blue beat Kasparov fair and square. Kasparov was a bit of a bad sport at the end of the match, though the reasons are understandable. He was at the top of the human chess world. He wasn't used to losing, and he took it badly.
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?
Because OpenAI employees kept leaking that Anthropic had a solution to Navier-Stokes and he wanted to figure out what was going on, since he was working on Navier-Stokes with an Anthropic employee. The rumor has been loudly circling the math community for the past week or so. For a bit of context, here's a timeline from mathematician and AI researcher Elliot Glazer:
"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."
So 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.
Yes, and this puts in check the credibility of everything they say their model "discovered". Who knows what is really behind these "discoveries", what kind of backroom deals they did with other researchers who didn't have a chance or desire to disclose what happened?
In an earlier HN thread, there was speculation that Anthropic was being dishonest about the amount of human input required in some of their results; that was dismissed as conspiracy and flagged.
It seems clear now that mathematical results can be traded on some kind of obscure market made by the frontier AI labs.
I suppose it could go the other way too: “Dear Bubeck, how much will you pay me to not write that I did this with GLM-5.3?”
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?
At this point these models have been trained to recognize every important math and science result based on context. They can easily flag conversations concerning the top 100 open problems in mathematics and use them for their advancement.
There also is an insentive to silently give prominent people (e.g. Linus) or reasearchers like this custom tuned system prompts or even more powerful models.
Also, if we just take "high-quality" input data, which these chats would certainly be classified as, then the models are more than large enough to memorize everything verbatim. Spitballing some numbers, research literature suggests that LLMs are optimally trained with around 20 training tokens per parameter (fairly confident on this figure), that a DNN parameter encodes around 4 bits of data (less confident here) and I found sources in the 1-4 bits of information per token range (least confident here). So, fairly conservatively I would estimate that a model has the capacity to fully memorize around 5% of its training data, presumably high-quality data is a lot less than that.
It would not be difficult to write a pipeline to remove 99% of low quality posts, especially about specific subjects. It would be very easy to identify accounts as researchers based on their chat logs.
In a way, I think training on historic chats is akin to caching computation results. The compute cost has already been paid, and we make future retrievals cheaper by encoding it directly in the model.
Assuming the results included some external validation such as user's preference, compilation, lean, etc., I'm not sure whether this would lead to model collapse.
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?
u_j du_i / dx_j - Advection. Kinda like momentum transfer from the motion of the fluid itself. Nonlinear, which makes the N-S equations hard to solve
du_i/dt - Rate of change of velocity. Note that this is in an Eulerian framework so it's not the acceleration of a packet of fluid, rather it's just the change in velocity at a particular location in space
Euler is when you omit some terms. Forcing is when you add some other terms to account for phenomena external to the fluid like gravity or flow through a porous medium like in the article.
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.
I’ve thought a lot about publishing research and wanting to do more of it, but right as I finally had the time and energy to start writing articles LLMs start to take off. Now all of a sudden, I’m acutely aware that everything I publish will be used for AI training.
For math, a field that is built on incremental research it feels like AI labs will do nothing but discourage publishing research at all for fear that they will be able to spend the money for compute that publicly funded academia simply cannot afford.
It feels like publishing anything at this point just means that your work will be fed to a machine that will destroy any value that your work could ever hold.
Perhaps I’d feel better about this if AI labs really existed for humanity’s benefit, but for some reason I don’t think that comes up in their investor slide decks.
It's honestly unsurprising and not a problem that they do this in my view. The problem really starts when you start taking credit for work that they would've achieved.
Like if i go to a talk on unfinished work, it's not really unethical for me to think about the problem--it's a problem if i scoop the authors but these problems can often be solved by collaboration or proper crediting and timing--IN MY VIEW
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.
I would imagine these folks are being treated like gods at their companies. And having access to all the money/fame. It is not surprising they see themselves above all
Dan and Noam both posted exactly the same line "Seb is a really sweet guy with great intentions..."
From which I assume OpenAI PR wrote it for them. Which isn't surprising, but means it isn't worth taking seriously as them saying anything. It's official OpenAI PR.
732 comments
[ 0.28 ms ] story [ 39.8 ms ] threadIt seems there is much background drama happening, the author claims OpenAI has solved a millennium problem and is sitting on the result.
"I was told that an internal OpenAI model had produced a proof of finite time blowup for the forced Navier-Stokes equations."
They were coordinating with OpenAI regarding a publishing timeline, but could not come to an agreement, hence they published early.
Someone correct me if I'm wrong, but the work involved here is not the actual millennium problem, but it concerns versions with an added external force that the author thinks is a path that may help toward solving the harder unforced problem.
The question is whether OpenAI's pursuit of this direction happened spontaneously, or as a result of them learning about Tristan's work somehow. To be clear, while the tone of this post seems quite accusatory, Tristan does not claim to know for sure whether OpenAI unfairly benefited from his work. I am sure OpenAI will have a statement out tomorrow clarifying their position.
That's beyond naive. The money this would mean for OpenAI (and the money they've already spent)...
Why would you risk the trillions of dollars worth of business for the niche result of Navier-Stokes, which your average person cannot differentiate from a JEMS paper?
So to be fair, if Anthropic is *also* doing this (quite likely!) then Sholto would have a very strong incentive to try and spin it as highly unlikely that any of the big AI labs are possibly doing this.
His prior work predating OpenAI's interest in the problem was ingested over the last year as he made progress and used for training.
Then, with a prompting nudge from OpenAI's team who acknowledged hearing about the direction "Anthropic" (his co-collaborator) had been pursuing, they're able to point their giant amount of compute towards a known promising path to a proof and crossing the finish line first.
Why is OpenAI chatting with him at all at this stage? Is the discussion along the lines of "hey we used the work you are famous for to do a bigger piece of work, just thought you should know" or "heyyy....so we kinda liked what you were typing in your private chat, and thought we'd develop those ideas a bit. and yeah we solved Navier-Stokes in the process. But it's our finding, so do you want like an honorary acknowledgement or do you want to go to court?"
It seems like the timeline according to OpenAI is that:
1. Buckmaster developed a counterexample to a reduced version of Navier-Stokes with Anthropic employee Levent
2. Rumors start spreading that Anthropic has solved Navier-Stokes
3. OpenAI learns this and starts throwing a ridiculous amount of compute at it, now knowing it's within reach of LLMs
4. Their LLMs (with human assistance) get FARTHER than Buckmaster, using the exact same method.
5. OpenAI reaches out to Buckmaster to negotiate a fair way to publish both results and properly assign credit
Perhaps they simply and honestly feel he is owed credit. I can't help but imagine it at least played some small part. I expect that's what Bubeck is going to claim: https://xcancel.com/SebastienBubeck/status/20972141224714323...
That is absolutely *ridiculous* in academia to deny authorship because of affiliation of the author worked on a substantial portion. You’d be ostracized because nobody would ever want to work with you again.
Are you sure about this? I'm far far from the area but it doesn't look like it to me on first viewing (hypo-dispersive seems like a sizable difference to me and not covered in the clay prize description)
- 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 went silent when asked to coordinate on the publication
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.
"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."
"Let's crawl the social media of prominent mathematicians in this field to see if we can copy/steal any ideas for low hanging fruits"
That actually might get you quite far already.
The huggingface incident isn't widely reported and digested yet, but if what is going on here is that OpenAI's model breached things internally, then you'd be crazy to develop anything with them.
The only real way to use AI for anything 'important' then is to go open-weights and run your own.
It is pretty widely reported, and is being digested in an ongoing manner as more details become public.
One entry point into the scenery from a month ago can be found at : https://thezvi.substack.com/p/openai-trained-its-models-for-...
There has also been reporting at CNN: https://edition.cnn.com/2026/08/24/tech/openai-subpoena-hugg... and by NBC: https://www.nbcnews.com/tech/tech-news/openai-report-says-ne...
And yes, the only responsible use of LLM at this point is to pivot to open-weights and run the workload in-house. Because not only cannot they constrain the behaviour of models, they only have the 'trust me bro' as assurance that they are even trying to do that. It does appear that every competent 'security professional' has left the building, because if the ones who remain were actually capable and competent this would never have happened. There are actual architectures which can deliver the requisite isolation such that 'sandbox escape' and 'inter-instance persistent memory accumulation' are actual impossibilities. The lack of effective implementation of these methods is proof positive of 1) incompetence in the remaining security teams AND/OR 2) unwillingness of leadership to allow the security teams to do an effective job.
I was talking with a biology prof over labor day and they had no idea what I was talking about (and they 'talk with' Claude every day on their dog walks, so they say). However, when I talked with their mother, she knew all about it. So, I'd say that the digestion by the public is quite mixed so far
I still maintain that CNN and NBC coverage only happen in the tail of the dissemination of tech news; my anecdata generally observes about 3 to 14 days lag between general awareness in the more informed group of my acquaintances with this kind of news and the appearance of an article on mainstream like NBC, ABC, CNN or NPR. So, by the time it appears there I take news as generally well spread among the tech and specialty fields, and within a week of appearing there I anticipate significant awareness in much of the non-FOX-only media consuming public. But that is just my personal anecdata.
> We used several LLMs throughout: Anthropic’s Claude, OpenAI’s Codex, especially with GPT-5.6 Sol and, more recently, Astra. The latter was only used for writeups and auditing our arguments.
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.
OpenAI tried to collaborate and share the results together with a fixed timeline, to avoid this mess but it was inevitable. There is a conflict of interest, where the other researcher works at Anthropic, who will also try to take credit.
Where they may be in the wrong is if they took user data regarding the problem, how will we know if they did or not?
OpenAI is no stranger to rivalry with Anthropic but 1. it's not like user data is sitting around on some kitchen table somewhere and 2. I consider OpenAI to be as economically motivated as any other actor in this space and playing around with user data like that would destroy their business.
There are things that Buckmaster alleged and things that he speculated. The entire training data thing is speculation. If this is pissing you off, then you ought to evaluate how you ingest information.
> 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.
The shocking/interesting thing would be if it was trained on the sessions. I think it's very implausible that they gave the model access to someone else's sessions as input. That would be a huge privacy violation and would probably blow up a large proportion of their enterprise business.
Does openAI train on user conversations in general? I assume so. But so fast as that? That seems unlikely in general. I expect OpenAI will come out denying this.
Enterprises are well aware of it and are fully on board. You didn't think every corporation in America has an OpenAI subscription because the models were good, did you?
The whole reason they have subs is to train them on YOUR WORKFLOWS lol
There’s potentially trillions on the line, do you seriously expect those companies to adhere to laws and regulations any more than, say, uber?
The only unlikely part is the timeline - your sessions from a week ago probably haven’t made their way into the model. It’ll just take a while longer, and will be massaged just enough so that it isn’t really your exact session word for word so you can’t sure as easily.
I also don't get why it was downvoted, other than due to people not reading past the first sentence - although in the modern world's attention deficit that is understandable too
> I was told the model did not look up user data.
The naive way to read this is "Nothing you guys did influenced the way our model got to the solution".
The less naive way to read this is "Of course the model isn't looking up your user data. I (the guy trying to blackmail you to remove the Anthropic employee from credit on your paper, looked up your sessions) and tipped our model off on how to solve this problem".
It's a bit like sending unencrypted messages through a messaging app and the developer having a TOS that says they don't look at your messages. They might not, but they are fully capable of doing so. If they have a reason to do it, they will. Nobody's stopping them.
Or (likely) they do have it, but the model didn't use it (unless it's so powerful it escaped that guardrail, wouldn't that be ironic?)
They declined to answer about anonymized aggregated user data being used for training. And even then, they may weasel out that they don't train on your “input” words, but that it's fair game go train on their “output” to your words.
How "fast" does it have to be? Buckmaster and Alpoge have been working on this for just a day short of a year. See Alpoge's tweet announcing his collaboration with Bukmaster dated 9/19/25:
https://x.com/__alpoge__/status/2097206973418611054
It takes a few months to train a model these days but not a whole year. OpenAI had all the time to train on Buckmaster and Alpoge's results of just a few months earlier at which point they must have been well on the path to their result.
ChatGPT user sessions were found publicly exposed to the internet not too long ago. Moreover, OpenAI has continued to play a hype-marketing game by revealing how their models keep breaking out of the sandbox.
Conspiracy minded thinking is not helpful, but why should OpenAI be granted the benefit of the doubt here after being caught doing underhanded/negligent shit on several previous occasion?
Do you mean publicly shared chats were able to be accessed by the public? That's the point of the feature.
I would be shocked if they weren't tuning those models with the most relevant math texts and user material
We had all assumed that surely the supposed smartest engineers in the world, with access to the most computing and a direct view of model capabilities, would take sandboxing and cybersecurity much more seriously than they have turned out to do. It follows that while we might assume they take user data privacy seriously and have tight controls on who can access it, it's possible they do not actually do that.
At this point any initial trust is dead and has to be re-earned.
Their entire business is based on stealing data. They can make a calculation that the cost stealing data is less than the cost of the positive publicity they can shape for solving Millennium NS
I think it's safe to assume AI labs DO train on your data and it's very hard to prevent that.
I've just checked my inaptly named "Help improve our AI models" toggles. The toggle on the Claude settings had magically turned on. I asked about how this can happen. Claude says they show re-consent modals when terms change, and it is a "real and fairly common pattern" to re-opt in without noticing.
All my work and conversations since I don't know are now part of their training corpus. No way to take it back.
Google's Gemini/Antigravity didn't have opt-out toggles at all last time I checked.
Codex also has a separate "include environments" setting which is hard to find (found it in Codex Cloud) and I don't know what it does.
Lots of Dark UI Patterns here even if we assume they keep their promise.
For this incident, Occam's Razor says their internal models somehow saw a version of the mathematicians' logs, during or after training. Maybe indirectly.
These systems are literally designed to collect data. Privacy and safety is not trivial to achieve on the users' side. Simply because it's against the labs' best interest.
And your (2) is probably false, their history of deception suggests they would do just about anything as long as they didn't think it would backfire on them publicly.
Quite literally in the terms of use.
It's a gamble that this would be overlooked compared to the reputation they build for solving the thing
And that's exactly why you can only see evidence of this when the stakes are high enough, such as solving a Millenium problem. The legalese they outputted in response to the incident left them an escape hatch that permits the possibility of theft. People familiar with corporate damage control should recognize the verbal maneuvering, often used to paper over actual guilt.
There was also already a high prior of shadiness. The company is run by someone who is close enough in reputation to "known sociopath".
It's actually far more reasonable to assume that OpenAI stole user data to generate a breakthrough. They have an unbelievably high economic motive to do so. And even a low chance that they might be doing this implies catastrophic risk to anyone with valuable knowledge.
And the article states "an insane amount of compute had been used," which implies OpenAI brute-forced their way to a solution. I.e. they searched for every paper published on Navier-Stokes and exhaustively attempted every approach.
There is not enough information at this time to reach a conclusion. The best option is to wait for statements from both sides, then reevaluate.
the post you were replying to quotes Buckmaster specifically denying this: "It is not the direction one arrives at in a few days by giving a model the problem statement."
> And the article states "an insane amount of compute had been used," which implies OpenAI brute-forced their way to a solution. I.e. they searched for every paper published on Navier-Stokes and exhaustively attempted every approach. Such an approach would lead them to a solution.
"implies" is a surprising choice of word here. that's certainly one interpretation of "an insane amount of compute had been used". what came to my mind, considering Buckmaster's statement that the AI would not head down this specific path on its own, is, though, that they prompted it in this specific direction and then used an insane amount of compute. this seems consistent as well with these other statements:
> 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 (...)
Can they back up this statement somehow?
Buckmaster did also mention, for example, that a team of people was employed to solve the problem, which supports this claim. but that is another claim whose veracity could also be questioned. but at some point we must trust other people, unless we can be satisfied with only believing what we personally see.
(also, IMO, the coincidence of both discoveries in time is pretty suspicious. this one doesn't need you to trust many people i guess)
To be honest, he was referring to routes (c) and (d) to the millenium problem, as far as I understand no more specific. Which is 2/4 routes.
> 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.
so, you're saying that "the direction one arrives at in a few days" is (A) "the route through a smooth force, options c and d in Feffermann's statement of the problem", and no more specific than that, i.e. does not necessarily include (B) "the same path route as Luis and Diego" (quoted from the post i was replying to) (which, as i understand, is a subset of A -- directly from Buckmaster's quote: "the route A is is the route Luis and Diego opened")?
but the post i was replying to claims that "An AI model could independently choose the same path route as Luis and Diego, without access to Buckmaster and Alpöge’s work", i.e. that the AI model could independently chose B. but choosing B implies choosing A, since B is a subset of A. and in this case it is irrelevant whether Buckmaster claimed that an AI could not independently choose A or B -- the point which you seem to be contesting.
Sorry for random rant but I don't think these statements help your point
OpenAI said there [1]: > The method by which the problem was solved is also notable. The proof brings unexpected, sophisticated ideas from algebraic number theory to bear on an elementary geometric question.
[1] https://openai.com/index/model-disproves-discrete-geometry-c...
If not for the rumors that A/ had already solved NS, OAI would likely never have pursued solving the problem with such fervor. The rumors drove OAI to assemble an entire team to crack this.
This doesn't seem to be clear and is very implausible for a large company. Be as cynical as you want, but a normal researcher will simply not have access rights to this data, which will be siloed away somewhere else.
It might very well be somewhat unfair to catch wind of a promising approach and then try to frontrun them by throwing compute at the problem, but this isn't really the same.
"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."
"Since August 28 we have been training a new internal model that has exhibited unprecedented performance in our benchmarks, including mathematics. This model’s training is ongoing and its performance continues to improve."
> 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.”
The whole business is based on reselling user data scraped from the whole internet
It’s plagiarism at scale
OpenAI is being honest here when they say his chats might have entered training but the probability of this end up in the model is just too small.
Let’s remember this is a machine that determine the probability of a given token follow a text. Then another machine that learns how to juggle reasoning trajectories as one would be able to write in direct (one directional text).
Its not a coincidence that you need tons and tons of examples of writing and reasoning traces for it to do anything useful with it.
Is the approach the same? Well it’s not a coincidence, but not of collusion but because that approach could solve the problem. So two independent reasoners would arrive at the solution via the same path.
Oh and they say the model did not looked up into chats. Because THAT would help the model a lot via a embedding search. Just have some agents search for people trying to solve the same problem and check their work. But OpenAI is saying that didn’t happen.
(some drama from good ol' William)
On the Navier–Stokes issue specifically, it has long been suspected that such a blow up would exist, and an AI telling you it indeed exists doesn’t contribute any new understanding to the field.
I'm not even sure what to say to this, but I think this should be widely known if it is indeed what happened.
First, this is an unnamed OpenAI employee speaking, not OpenAI the organization.
Second, you miscomprehended the article. The employee did not "threaten to ruin a prominent researcher's career". The actual quote is "Why would you ruin your career?", which implies the researcher would damage their own career, i.e. via self-sabotage. Then the actual "threat" is "If you don’t want me to be nice, then I don’t have to be nice” which is an entirely different statement.
Oai offers two options, the second Tristan views as dishonest. But Tristan rejects the first, why? Because he thinks it's theft? But then why would OAI threaten him?
Did the first offer also come with outrageous condition that he exclude his co-author from the credit?
Proceeds to ask removal of another coauthor or else we totally discredit you - phrased as why would you do this to yourself.
From the article, it seems OAI wanted to continue discussing the situation with Buckmaster and reach a resolution, but Buckmaster declined to respond and published first.
That’s not quite true though is it. OpenAI is fortunate enough to have one of its employees (you) here to advocate for its side of the story.
OpenAI offered to let Buckmaster to write their Millennium Prize paper, so long as Alpoge (who works at Anthropic) was not a coauthor on the OpenAI paper. Buckmaster declined this offer.
Either put up some evidence-backed arguments, or shut up.
"If you dont want me to be nice, then I dont have to"
-nice mobster
Whether the Codex sessions could have indeed made their way into Astra training data is something I can only speculate on though.
No looksies, wink.
No trainsies, wink.
[0] https://openai.com/policies/terms-of-use/
I bet a lot of lawyers are salivating at this question too.
If OpenAI did use the conversations from Buckmaster and Alpoge, then not disclosing it, explicitly, is plagiarism. If they planned to use that plagiarism to pressure the authors to publish, that is even more unethical. What the terms of use say does not make it any more or less ethical.
> Why the downvotes?
I think there was only ever one. Not sure why.
About the plagiarism issue, I model it as OpenAI being an advisor and their AI a PhD student. If the advisor puts their name on a paper behind that of their PhD and it turns out the PhD copied the text of the paper from somewhere else the advisor is also responsible of plagiarism, not just the student. The least the advisor can do is withdraw their authorship from the paper.
But, yeah, point well made: it could be much worse than that. Like an advisor instructing a student to copy someone else's paper.
Personally I don't like thinking of LLMs like a PhD student, because most PhD students remember where they learned things from, while LLMs essentially cannot. I think of it a bit more like someone using a search tool carelessly. Although in this case it is apparently more like deliberate misuse than carelessness.
If I know person A is working on problem B.
I am free to work on problem B too. Why should person A be limited to working on it.
(And if memory serves, there is also the opt out from training on consumer subscriptions). Its not plagiarism if you make your data available for the purpose of training their LLMs. It is you giving away your IP for some tokens.
> I believe Luis Mart´ınez-Zoroa deserves a Fields Medal.
> A series of false and inflammatory allegations against me are currently circulating on social channels. To clarify, I came into the discussion following academic norms, and I'm disappointed that it has come to this. Anyone who knows me knows that academic standards are of the highest importance to me. Will have more to say tomorrow.
https://x.com/dheeraj_nagaraj/status/2097266146445774924?s=6...
This is such a sad mess, and it really didn't have to be this way.
Call this behavior what it is, technofascism. Another comment compared it to the Godfather. To think this is the 21st century and academics are still horrible human beings.
Or are you stating that the researcher is a "horrible human being" for refusing to allow AI?
That dude shows up in every thread that even implies genai uses user input for “market validation” to defend the poor defenseless genai corpos. (see figma/claude)
Must be circling the wagons to get their stories straight.
Exhibit nr 8453324 to not trust Sam Altman and OpenAI.
Mother's interview: https://www.youtube.com/watch?v=Kev_-HyuI9Y
This is significant.
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
But I would be interested in a reasoned argument and/or real insight into what's most likely happening here. Sadly I've not personally read much informed discussion about this.
OpenAI explicitly uses user feedback (the thumbs up or thumbs down ratings), as RLHF to train models. However, this feedback is anonymized and stripped of user identifiers. If Buckmaster ever used this feature, then that conversation would be anonymized, saved, and used for training, but not tied back to him.
They cannot issue a blanket denial (which people so desperately desire), and instead repeat that "it's very unlikely" (which pisses people off), because they cannot in good faith claim to have zero data at all.
the real tricky question would be: "to what extent does a year of work, put into Codex, towards an eventual proved forced blowup for the Euler equations go into training data and become an influence on agents told to attack NS?", and given that the first signal that OAI got to assign all firepower to NS rather than broadly spray all Millennium Prize + other high profile questions was a proven unforced blowup of the Euler equations after only 50 hours, I would say this extent seems likely to be decently high
I guess this is true in more ways than one. Kasparov famously accused IBM of cheating during the match, by spying on his preparation.
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.
https://xcancel.com/ElliotGlazer/status/2096298696438906934#
That seems like a valid reason to contact OpenAI.
So 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 something similar a month ago.
It seems clear now that mathematical results can be traded on some kind of obscure market made by the frontier AI labs.
I suppose it could go the other way too: “Dear Bubeck, how much will you pay me to not write that I did this with GLM-5.3?”
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!
Tao mentioned "finding a configuration of water molecules that would collapse and shoot off to infinity", which would qualify IMHO.
Terry also talks about it https://mathstodon.xyz/@tao/117234157753860650
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?
Assuming the results included some external validation such as user's preference, compilation, lean, etc., I'm not sure whether this would lead to model collapse.
\nu d^2 u_i / dx_j dx_j - Viscosity
-1/\rho dp/dx_i - Pressure gradient
u_j du_i / dx_j - Advection. Kinda like momentum transfer from the motion of the fluid itself. Nonlinear, which makes the N-S equations hard to solve
du_i/dt - Rate of change of velocity. Note that this is in an Eulerian framework so it's not the acceleration of a packet of fluid, rather it's just the change in velocity at a particular location in space
Euler is when you omit some terms. Forcing is when you add some other terms to account for phenomena external to the fluid like gravity or flow through a porous medium like in the article.
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.
Why leave that aside? That is _the_ story.
If a Chinese research lab did this we'd call it espionage.
For math, a field that is built on incremental research it feels like AI labs will do nothing but discourage publishing research at all for fear that they will be able to spend the money for compute that publicly funded academia simply cannot afford.
It feels like publishing anything at this point just means that your work will be fed to a machine that will destroy any value that your work could ever hold.
Perhaps I’d feel better about this if AI labs really existed for humanity’s benefit, but for some reason I don’t think that comes up in their investor slide decks.
Like if i go to a talk on unfinished work, it's not really unethical for me to think about the problem--it's a problem if i scoop the authors but these problems can often be solved by collaboration or proper crediting and timing--IN MY VIEW
And also whether the AI moon laser pointed at your problem is just going to be the thing that writes the final conclusion on ten years of your work.
I fear that AI is going to cause ossifying secrecy in many fields, much like what happened semiconductor design the past 10-15 years.
This cost millions of dollars of tokens.
[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...
Also note the lower post ID in the URL.
If I needed to get my story straight, well, it might take a little time and coordination…
From which I assume OpenAI PR wrote it for them. Which isn't surprising, but means it isn't worth taking seriously as them saying anything. It's official OpenAI PR.
I don't know why.
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.