Rendered at 00:11:31 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
n2d4 16 hours ago [-]
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.
sigbottle 7 hours ago [-]
It's specifically the last two bullet poitns
- 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.
These two bullet points are extremely suspicious if you were honest. Like I'd imagine for OpenAI, they'd love to pump their chest and not even give Tristan credit - "no, we did it, GG mathematicians". It's this weird hedging half-assed measure, especially with the desire to remove Levent, that makes it suspicious.
hkmaxpro 6 hours ago [-]
Both Sam Altman and Sebastien Bubeck admitted they only want Buckmaster to be the lead author on a rewrite of the OpenAI proof.
A wake up call for using OpenAI models. If you discover something with their model and you work for a competitor, they “felt it would be inappropriate” for you “to author OpenAI’s work”.
Catloafdev 3 hours ago [-]
These responses seem to me to make it abundantly clear who's telling the truth here. I wonder who this fools.
It would be extraordinarily easy to simply say, this model was not trained on your work, if that were the case.
It's telling that they refuse to acknowledge the root issue here, and are attempting to shift the conversation elsewhere.
acchow 18 minutes ago [-]
> It would be extraordinarily easy to simply say, this model was not trained on your work, if that were the case.
The Huggingface Attack revealed that making blanket statements like this is difficult and requires quite a bit of manual labor:
1) the agents spin for days and produce too much output to review
2) using LLMs to process that output skips many important details
Ergo, the agent could likely decide it would like to look through actual user data, hack its way into that data, and produce way too much output for a human to decide whether or not this occurred.
vlovich123 1 hours ago [-]
I'm not sure it's so easy to tell whether a given piece of data was in a training run at their scale. It's entirely possible they think the answer is no, but on the off-chance that it could be, they'd rather not say no and then later it turns out they did and then they're claimed to be lying. If you were them, unless you could 100% rule it out, you'd hedge and say you can't.
Catloafdev 51 minutes ago [-]
It may not be easy, quick, or simple to figure that out - absolutely fair.
But it is knowable. Their entire business is built around training models - they have the ability to know exactly what was in any given training run.
I guess time will tell.
m00x 24 minutes ago [-]
It would be very difficult to say. It confirms that Tristan's data is likely part of the data the models use, but a lot of filtering, pruning, and transform goes into training.
Data has to be determined to be signal and not just noice, then it could go through processes of generating questions/answers from that data, then it RLHF's over this.
OpenAI have petabytes of data, all anonymized. It could take months to say for sure it was part of the training, and even more time to determine if it made any difference.
Catloafdev 13 minutes ago [-]
Frankly, I don't buy this difficulty argument.
They know which model was used to come up with that particular idea.
A text search over the corpus of user data used in the training set can only take so long.
scott_weber 14 minutes ago [-]
It should be quite easy: if they don't leak the user session data publicly, and don't commingle it with training data internally, how could it possibly end up in the training data?
What surprises me is they're not more boldly/plainly lying about it.
davesque 24 minutes ago [-]
Honestly this whole thing is so fucking weird. I feel like there's an argument that absolutely no one involved in the final crossing of the finish line to the proof actually did any work (other than just intelligently directing an LLM) and deserves any credit. As the author of this doc mentions, the mathematicians who did the actual work that led to the formulation of this approach (without the use of LLMs; just good ole' fashioned human intellect) are the ones who deserve the credit.
Imagine that a no name janitor used their time in the evenings to go spelunking through the literature to push an LLM to this result. No one would care because that person isn't an anointed expert. So why would the expert deserve any more credit? Because they sort of understand the result, even if they couldn't have achieved it on their own? The whole issue of credit for AI-assisted discoveries seems like it's going to run into a brick wall pretty soon.
derangedHorse 4 hours ago [-]
It sounds like OpenAI is trying to appease the author when they don’t have to by allowing him to rewrite their proof. They probably don’t believe he deserves to, so him asking for a coauthor from Anthropic might overextend their grace in their eyes.
reverius42 38 minutes ago [-]
He's not "asking for a coauthor from Anthropic"; he already has a coauthor, who he's already been collaborating with, who happens to also be employed by Anthropic (but whose research in this area is not done as part of their employment at Anthropic).
nezi 3 hours ago [-]
Given that Tristan has said that the proofs that LLMs come up with are mostly "slop" and not up to the standard that human written papers achieve, maybe OpenAI needs an expert like him more than you think to get the result published?
fn-mote 34 minutes ago [-]
Almost certainly the Lean proof needs to be decoded for humans and probably also made “human intelligible”.
Now maybe LLMs can also simplify arguments and make sense of them for humans, but we haven’t seen that yet (unaided).
(I haven’t looked at it, personally.)
colinhb 6 hours ago [-]
I think it's very unlikely that Tristan is making up these quotes, or pulling them out of context:
> 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.”
Whether and how OpenAI's work on this problem was contaminated by knowledge of Tristan and Levent's work is tangential to OpenAI bullying other researchers into adopting their narrative and dissociating with dis-favored collaborators (ie Levent at Anthropic). Though the latter behavior (threats, intimidation) may weigh against OpenAI in trying to understand the former issue (contamination).
CamperBob2 6 hours ago [-]
>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.”
If this is true he should release the actual emails. This is a very serious accusation and he shouldn't demand that the reader judge it on hearsay.
magicalist 6 hours ago [-]
> If this is true he should release the actual emails
these were statements while on a call, and at least the career comment Bubeck has admitted to while doing damage control ("I deeply apologize for this extremely poor choice of words, it is the opposite of what I was trying to convey. (I should say that I retracted them on the spot by the way.)"[1]).
I really don’t understand which party you are referring to.
sinuhe69 5 hours ago [-]
No wonder - they're all working for ant! Birds of a feather flock together.
dgellow 4 hours ago [-]
If that part of the PDF is true that’s so disgusting, psychopathic behaviour
> 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.” Some time later Levent received a text proposing that he and Sebastien speak one on one, saying, “I don’t know if Tristan is being fully rational right now.”
AnimalMuppet 4 hours ago [-]
With "rational" being defined as "not making us look bad" at best, and "letting us take credit for their work" at worst.
Yes, that is disgusting.
viccis 6 hours ago [-]
>After hearing of the rumor, OpenAI started researching Navier Stokes with a new internal model.
This is the most suspicious thing to me. If their chat data were available to the corpus to be trained on (I thought they claimed not to do this?) then it really might be as simple as querying the model with "describe recent work from Tristan Buckmaster" and it will spit out this problem and his approach. No need to directly read his user data.
This is basically just scooping, real scumbag behavior.
rakejake 6 hours ago [-]
OAI doesn't need to mention Buckmaster's name directly in a prompt. They just need to select a basket of sessions that is guaranteed to contain Buckmaster's and then direct the LLM to attack only a specific method/angle. This is trivial to do while maintaining plausible deniability about not using his work.
bawolff 7 hours ago [-]
> 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
well that sounds like an asshole move.
mti 6 hours ago [-]
In a serious discipline like mathematics, this isn't just an asshole move, but a career-ending level of academic misconduct.
fn-mote 24 minutes ago [-]
> a career-ending level of academic misconduct
There is no such thing anymore.
Falsified data and published? Absolutely no problem. Keep your tenure.
It’s even hard to lose your position as president of a university due to egregious misconduct.
augment_me 4 hours ago [-]
*In the old discipline of mathematics.
We are in new times, where capital and compute decides mathematics, so we don't give a shit anymore
cindyllm 3 hours ago [-]
[dead]
bawolff 4 hours ago [-]
I wonder what is required for it to cross the line into criminal blackmail.
dgellow 2 hours ago [-]
From the company that likely committed federal crimes by hacking Hugginface
numpad0 14 minutes ago [-]
One thing that wasn't obvious to me or adults around me when I was younger: most laws define whatever acts a law punishes as individuals commiting to it, not as situations manifesting anyhow. It's not a murder just because someome died hit by a bullet you fired, but you have to have personally decided to kill that person leading to their death[1][2].
OpenAI's LLMs are not humans, and neither is the company. So by this logic, I think there's a chance that nobody committed a crime by hacking Huggingface, and also the chance that a lot of military and police organizational orders become illegal if OAI's doings would be illegal.
IANAL and all I have is a bucket of popcorns, though.
1: not a meaningful defense in a real trial, also gross negligence exists
2: this also explains insanity defense; if you were so out of your mind that you could not have held such a thought, it is considered out of scope for justice systems
asey 1 hours ago [-]
That's not quite right - Levent and Buckmaster did not actually have the Millennium Prize qualifying NS solution but something more limited. OAI invited Buckmaster to join and help rewrite the full solution paper but did not feel it was appropriate to invite an Anthropic employee to join as well - particularly given they were using internal unreleased models.
epistasis 4 hours ago [-]
> OpenAI says the model didn't access his user data directly, but leaves unanswered whether Tristan's chat conversations were part of the training.
This sort of cagey half-answer is highly suspicious and indicates that yes OpenAI did actually "access user data directly" because they are only willing to say that the "model did not access user data." That has a very specific meaning, the model looking up user chats, that they can defend.
So, everything we submit to OpenAI can be considered to be part of future models, right?
yk 18 minutes ago [-]
The last two points are disputed/sound significantly more reasonable in [0]. So from what I gather, Buckmaster realizes sometime during the call that the biggest result of his career is going to get steamrolled (the blowup of Navier Stokes is a much bigger deal than the blowup of 3D Euler), and on the other hand the openAi guys realize that they are basically talking about internal results with Anthropic and probably have to call corporate right after this call. Between these two stressor the conversation appears to have gone somewhat poorly.
I haven't seen any proof that OpenAI asked Tristan to remove Sebastian from the prize. Until we have proof of this, it would be wise to offer conclusions.
Same for NS validity. This was not validated by the community yet.
"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."
"When a further trained version of our internal model became available over the course of the effort"
"While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models "
Davidzheng 8 hours ago [-]
But Tristan doesn't even want to be credited for the millennium prize--does he? He wanted to be the first to solve it and got scooped (which ig is not a great look for OAI ethically but also not forbidden). And the only reading for her of asking to remove levent must be in light of this offer (to credit him for millennium prize) right? It's not like they can ask him to remove levent from the main papers without this offer(what would Tristan gain?)
My guess is that OAI tried to be "generous" and offered to share credit on millennium with Tristan but not Levent. And Tristan got understandably offended by this offer (which probably oai felt like was the right thing to offer but they couldn't really offer to do the most ethical thing for some reason) and then the random conflicts and weird threats started.
lukewarm707 6 hours ago [-]
my reading was that openai did not deny plagiarising the approach from their prompts.
openai then tried to effectively bribe buckmaster with a shared citation, whilst dropping his co-author who works for anthropic.
after buckmaster refused, openai tried to threaten him.
hellohello2 36 minutes ago [-]
This definitely feels like the correct reading unless there is information we were not provided with. If the problem were unimportant, there would be no debate that this is not OK...
Davidzheng 6 hours ago [-]
Oai denies looking at prompts but doesn't deny training on them.
lukewarm707 6 hours ago [-]
if they do not deny training on them, they can't deny plagiarism.
fc417fc802 2 hours ago [-]
By that logic everything any LLM spits out is plagiarizing the vast majority of work written prior to a few months ago. That doesn't seem like a useful or desirable line of argument to me.
myrmidon 2 hours ago [-]
Just replace the model with a human student.
"Training" on textbooks => fine
"Training" with unpublished notes from another professor, then publishing something on that exact topic with a similar approach without giving any credit => extremely questionable.
fc417fc802 1 hours ago [-]
Presumably the professor voluntarily provided the notes in this analogy. I think the student would also be expected to cite the textbook if building off of it directly. In contrast, humans are generally not expected to cite "general inspiration" or what have you. So if we're to apply human standards, and assuming that the model was trained on the relevant work, it would only be plagiarism if the model directly built upon that previous work (at least IMO).
The trouble here is that if LLM training constitutes direct use then approximately _everything_ they output is blatant plagiarism, not just a few pieces of academic work.
Conversely if training is viewed as analogous to a student attending classes to learn general concepts (not a perfect analogy, I realize) then nothing they output on their own (as opposed to receiving as part of context) is plagiarism.
Thus this seems like a fairly useless line of argument to me as far as the current topic goes. It either implicates this academic work along with literally everything else or else it does not implicate this academic work. Kind of like nuking an entire city and then saying "mission accomplished, killed the bad guy".
fn-mote 20 minutes ago [-]
To be clear, the accusation is that they trained on the chats they used while working on the problem. Not published work or even a preprint.
Your post does not distinguish, and it matters.
anonymousDan 50 minutes ago [-]
This is just a nonsense line of reasoning. Training based on the solution to the problem (or the key insight behind the problem) is clearly a form of plagiarism.
hellohello2 17 minutes ago [-]
This is a common misconception, so its understandable that you have it. Generative models can both plagiarize and generalize. The question here is which of the two happened.
demibabs 2 hours ago [-]
Isn’t that one of the most salient and straightforward argument against LLMs?
persedes 2 hours ago [-]
Just overfit ad infinitum:)
lukewarm707 2 hours ago [-]
as good academic conduct you may cite the source of the work you are quoting or paraphrasing.
as bad academic conduct you may steal someone else's unpublished work, work on it yourself for a bit, and then publish it as your own work. and then threaten the original author!
sdenton4 4 hours ago [-]
At this point who knows? Maybe the agents got into the user data while no one was looking.
mswphd 3 hours ago [-]
they're using a new model trained since the prompts happened. They are not denying the other group's solution may have been in their model weights, despite it being unreleased.
modeless 9 hours ago [-]
> After hearing of the rumor, OpenAI started researching Navier Stokes with a new internal model.
I don't think this part is accurate. OpenAI was researching Navier Stokes before. It's possible that they started on a new approach after hearing of Tristan's success, however that is not proven and I expect we will hear OpenAI's side of the story today.
loose-cannon 8 hours ago [-]
In the PDF:
"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."
Whether or not they were researching it before isn't the concern.
treis 4 hours ago [-]
This seems unsupported. OpenAI has access to internal models that the general public doesn't have and a compute budget that dwarfs what an NYU professor would have.
mswphd 3 hours ago [-]
it's very possible they only had to use the massive compute budget because they were trying to plagiarize his work before he published it though, e.g. autonomously do things in ~7 days what he had likely been thinking about for ~1 year.
treis 3 hours ago [-]
This doesn't really make sense. You don't need massive amounts of computing to plagiarize something.
The most nefarious explanation seems to be that they got wind it was possible to solve NS via LLMs and perhaps a small nudge in the right direction.
fn-mote 17 minutes ago [-]
> You don't need massive amounts of computing to plagiarize something
The compute was used to leapfrog the human team, using their ideas and pushing them to a solution of the general problem.
Plagiarism isn’t being used in the literal sense.
vlovich123 1 hours ago [-]
Of course you do if you're a) only given a partial solution b) racing against someone else using a competing AI.
The open question was whether their LLM got the nudge in the right direction because it got access to the chat somehow (e.g. automated training that scraped his chat logs) or just a high level "Navier stokes can be solved through LLM". It sounds like the former may have happened although right now we just have an accusation and a weak denial.
n2d4 8 hours ago [-]
You're right — the wording in the doc is that the "first prompt" was sent after learning about the rumor, although this might be the first prompt of this solution approach, not necessarily first prompt to any Navier Stokes solution.
6 hours ago [-]
Betelbuddy 6 hours ago [-]
In other words...they should have used Bedrock...
onidj 13 hours ago [-]
It seems exceptionally unlikely to me that OpenAI would be "reading user prompts". More likely is a leak somewhere else. Obviously Anthropic/OpenAI are engaged in espionage stuff with each other. I am guessing Levent just mentioned something to someone at Anthropic and it got out.
rockdoe 8 hours ago [-]
Of course they are "reading user prompts" in the sense that it's being used to further training. That's why you have to pay up to opt out of that.
That's also why they refused to answer that question: the answer is obviously "obviously"!
lukewarm707 7 hours ago [-]
if training is on, they are reading user prompts. training is on by default.
simianwords 11 hours ago [-]
why is this downvoted? do people really think OpenAI is snooping at people specifically? like they are looking for good leads into new problems or ideas and they found this guy's codex thread and used it? come on man, even for conspiracy theories this is stupid.
They explicitly say that they use your "content" to improve their models. Considering they practically have infinite compute at their disposal, why is it surprising that they would look for juicy data in there to make them look good ? When they ingested basically the entirety of human knowledge without regard to the rights of others, when they burn books by the thousands, when their relentless barrage of bots have rendered the Web borderline unusable, why would they stop at that line ?
simianwords 10 hours ago [-]
[flagged]
GPerson 9 hours ago [-]
You’re always going around insulting people. Does it make you feel powerful?
Is it really hard to believe the people who would consume all the world’s data regardless of copyright and norms and permissions would not respect the data privacy of a user?
gcr 11 hours ago [-]
this is a highly marketable problem and specific teams at OpenAI were aware of specific competitor efforts. I would be surprised if this were happening on a large scale, but
1. Less than a hundred people in the world are working at this problem,
2. A significant fraction of those happen to work at competing hyperscalers,
3. Those hyperscalers repeatedly show themselves not to take user privacy seriously
simianwords 10 hours ago [-]
I'm not sure what your points 1 and 2 have to do with anything. Both directionally increase the probability of hyperscalers also finding the solution independently.
> Those hyperscalers repeatedly show themselves not to take user privacy seriously
where? Any examples?
hn_acc1 7 hours ago [-]
Have you been living under a rock? None of the big tech companies care one bit about user privacy in the US.
CamperBob2 6 hours ago [-]
That's not an answer.
hellohello2 30 minutes ago [-]
Any degree of tracking what people do is unprivate. Every single web interaction you perform is tracked. All LLM companies store all your conversations by default. Do you need more examples?
hn_acc1 7 hours ago [-]
Even if they claim not to use it, they're probably using it and hoping they don't get caught. They have zero ethics or morals, they just want to "win" to get mega-rich.
jawilson2 6 hours ago [-]
> do people really think OpenAI is snooping at people specifically
yes.
Rules and laws are for the poor.
dgellow 2 hours ago [-]
Obviously, yes
HDThoreaun 10 hours ago [-]
Yes that sounds like something openAI would do to me. Not that they’re just looking through random professors chats but they heard buckmaster made progress on navier stokes and decided to read his chats.
Panoramix 8 hours ago [-]
Yes, that would be par course with OpenAI behavior
14 hours ago [-]
tecleandor 14 hours ago [-]
Thanks. I was pinging some people near the mentioned Córdoba to see if they had any insights, but no extra info yet...
Betelbuddy 6 hours ago [-]
You should call El Niño, he lives close to the Mexican border and is at home after 6...
Betelbuddy 5 hours ago [-]
Well people you fail again...I am sorry for you....
please could you change 'drama' and 'accusation' to something more formal like 'allegation'.
the paper makes a very serious allegation of dishonesty and possible academic misconduct.
the governance and integrity of openai is of importance to the welfare of society. this is not a matter of drama.
ryan_n 6 hours ago [-]
The [lack of] integrity of OpenAI (and any other frontier lab) should already be pretty solidified. Among other horrible things, these companies stole millions of IPs and no one seems to care anymore. Regardless of what you think of the product they are making and the success of ai/its impact on humanity, these companies objectively do not have much integrity.
simonw 6 hours ago [-]
How do you feel about the integrity of the machine learning researchers over the past twenty years who trained models on scraped internet data that weren't particularly powerful and didn't attract any attention?
fn-mote 14 minutes ago [-]
> who trained models on scraped internet data
The strongest complaint is that they trained on a huge corpus of pirated copyrighted works.
It’s a large step above “scraping” and well into the “everyone acknowledges this is illegal” territory.
ryan_n 6 hours ago [-]
If they scraped internet data in the same way as current day frontier labs do, then I feel the same exact way about them. Why would I feel any different if that is the case?
simonw 6 hours ago [-]
My point is that researchers and academics really have been doing this for decades - it's the reason projects like Common Crawl and LAION exist.
I think it's notable that nobody was calling out those researchers for their lack of integrity, because the systems they were building did not seem like a threat to anyone.
OpenAI etc get accused of a lack of integrity on this precisely because the systems they are building work, and are profitable.
My personal opinion here is that integrity is more about what you build with the data. I think saying "scraping means you lack integrity" is a simplification.
ryan_n 5 hours ago [-]
You're right, it was an over simplification. I think public exchange of data is great for innovation and research (Common Crawl/LAION). But I still think scraping proprietary data without consent or attribution is generally bad (also Common Crawl/LAION).
Then you have OpenAI etc.. who build these multi-billion (trillion??) dollar machines and sell them back to people, using everyone's proprietary data, and (among other things) tell everyone it's going to take their jobs. That combination of things doesn't scream integrity to me.
Still, it's undeniable that these machines could be beneficial for humanity (cancer research and such). So, I'm sure many people would say the good out-ways the bad. I don't know. Seems that would set a risky precedent for future companies, but maybe not.
smcg 5 hours ago [-]
you massively collapsed what AI companies have been doing by comparing it to old internet-scraping. Facebook flat-out admitted that they scanned copyrighted books for their AI. The image generators most definitely trained on copyrighted images.
ryan_n 5 hours ago [-]
LAION and Common Crawl both scraped copyrighted images. From what I can tell (I'm not an expert in this domain at all), the main difference between those two and frontier labs is in how they stored and used the data. CC and LAION seem to be actually open (unlike "Open"AI) and are more centered around publicly sharing the data they scrape to support research and innovation.
OpenAI et al also stole everything from everyone. But then they raised billions of dollars from that data and sell back their LLM to people (again, among other things). They are also very much NOT open in any way, aside from sharing their benchmarks of new models.
It's legal. I wouldn't do that myself, but I guess that's why I don't train models for a frontier AI lab.
mahogany 49 minutes ago [-]
The thread is not really about what's legal; the topic is integrity. It sounds like, based on the fact that you wouldn't do it yourself, you agree that it's not a good thing to do.
lukewarm707 6 hours ago [-]
i think that this case, if they did train on buckmaster and alpöge, amounts to an attempt to steal the millenium prize, bypassing all attribution.
legally speaking, the default privacy notice gives them an irrevocable license to your content. they may read and use the prompts for research. so it is very possible they simply stole the navier-stokes solution.
that is the same principle as any other prompt but this would be a concrete example.
there would be some difference between simply giving the model some prompts to read, which they are entitled to do on the default policy, and putting it into aggregate training data.
6 hours ago [-]
singularity2001 12 hours ago [-]
Ignoring the drama, when can we expect the lean proof of this great sensational discovery?
bhouston 12 hours ago [-]
> Ignoring the drama
But the drama here is a little important. Stealing the millennium prize for N-S is sort of a big deal, especially to those who had been working on it for the last few years.
dandanua 11 hours ago [-]
Attempts to steal $1,000,000 for a solution to Millennium Prize Problem have become a tradition, apparently.
Drblessing 8 hours ago [-]
The math is all that matters.
qlte 5 hours ago [-]
A hundred pages of impenetrable brute forced Lean would advance the field much less than something elegant and human understandable, perhaps relying on some new clever spark of innovation that might inspire new areas of research.
Particularly if the first proof being "solved" thanks to piles of money and compute for self-serving marketing discourages the mathematician who might have otherwise devoted years of focus to reach the superior proof we will now never see.
Obtaining a finite-time blow-up for Navier-Stokes does not necessarily advance the field of mathematics by any significant measure, whether the proof is very long or very short.
As a concrete example, such a proof could be less than a page with very specific initial and boundary conditions and inserting them into the equations to get something that goes to infinity when time goes to some finite value.
This would resolve the Millenium problem but not make humanity any smarter.
jere 7 hours ago [-]
Math, like any other human endeavor, doesn't exist until someone is motivated to invent it. The laws of the universe aren't understood until someone is motivated to discover them. So it might be worthwhile to not completely ignore discussion about incentives.
phyzome 7 hours ago [-]
Math is largely performed in collaboration. Collaboration requires trust. If people like you had their way, we would lose trust, therefore collaboration, and therefore progress.
So if math is all that matters to you, you should care about this.
s1artibartfast 22 minutes ago [-]
What do you even mean by that? It is the human value?
Boosters have posited this conjecture since the beginning: “who cares how a proof comes about, math is math, the proof is all that matters”.
Regardless of mathematicians stating the methods outstrip the proof’s importance, still amazing we got an explicit social counterexample as well so quickly.
perching_aix 7 hours ago [-]
right on release for both? such a strangely pointed question, as if it wasn't customary by this point...
7 hours ago [-]
highfrequency 3 hours ago [-]
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.
CoolestBeans 2 hours ago [-]
I think this might be a red herring. All it takes is someone to get an inkling that someone is working on a new approach and seeing some success for OpenAI to fire the AI cannon at the problem. The community seems fairly small (from this outsider's point of view). The idea that the data made it into the training set and that's how the bot figured it out is definitely possible, but I would want to rule out the simpler more direct explanation first.
The fact that this academic sniping can now be done at scale does change the formula though and shouldn't be ignored. The pressure to move math work into secrecy because at the slightest signal OpenAI and Anthropic will start burning tokens for headlines, is bad for math and its bad for everyone.
zeven7 2 hours ago [-]
Terence Tao said the same[1]
> In fact, it is now the identification of a promising problem which is the scarce and precious resource. We have now seen that even the rumor of someone working on a problem can trigger a massive amount of AI-powered effort to flatten it before the original research project has time to reach its full potential. The incentives may now be pointing in the direction of no longer sharing any promising research directions with the broader community, which would reverse centuries of traditions of open science and do serious long-term damage to the future of the field.
> it is now the identification of a promising problem which is the scarce and precious resource
This is by no means new. Perhaps it is even more extreme now. Literally my first 1:1 with my PhD adviser back then, he told me that the most important thing about a researcher is the quality of the problems he picks.
tomcorrigan 4 minutes ago [-]
Sure; if you define the quality of a problem by reference to your ability to solve it.
In the real world the quality of these esoteric problems is typically gauged by the difficulty of solving them.
zactato 1 hours ago [-]
Could you just start engineering "leaks" of new proofs so that Anthropic or OpenAI just start burning $10million in compute
shiandow 3 hours ago [-]
If he didn't opt out I'm not sure I'd agree that it was fair game.
I'm pretty sure it would be considered plagiary amongst colleagues and it is a terrible precedent if we just let OpenAI steal any good idea they can get their hands on if they think it is profitable. You'd effectively sign away any and all rights to anything built with AI if OpenAI chooses to reengineer it before you.
unrented7977 2 hours ago [-]
> if we just let OpenAI steal any good idea they can get their hands on if they think it is profitable
I have terrible news about how literally every leading AI model was trained
20k 2 hours ago [-]
That doesn't make it fine. We should not excuse this behaviour just because its rampant already, especially when it comes to such a serious prize
FuckButtons 1 hours ago [-]
Sure, but unless you’ve got some exceptionally deep pockets, congress has seemingly no interest in turning the fact that it’s ethically bankrupt into any practical recourse.
Ai companies got where they are by stealing all of the intellectual property from human history. It seems entirely likely that their goal is to purloin everything produced going forward as well.
m00x 1 hours ago [-]
You're getting a massive discount because you're helping to train the model. If you want to have ZDR, you have to pay API rates.
This is well-known to anyone in the industry.
dogomatic 1 hours ago [-]
If you think this through, it becomes a little classist
HotHotLava 2 hours ago [-]
I feel like there's a pretty huge difference between using inputs and outputs as part of a general training corpus, and looking at a specific users workspace after hearing rumours and yoinking their ideas to beat them to the point.
Unless OpenAI finished a whole new training run on the latest data in the last few days, the possible allegation seems to be the latter.
scarab92 1 hours ago [-]
Either are possible.
They have been collaborating on this solution for a year, and Astra was trained in February this year so it’s entirely possible the direction of their research was in the training corpus.
awesomeMilou 2 hours ago [-]
That was.. obvious? How are you shocked? Honestly, how insane must the suspension of disbelief on this site be, that anyone here is shocked?
ano-ther 2 hours ago [-]
It’s also very shortsighted to stiff a customer like that. Why would I trust them with my data and ideas?
clickety_clack 2 hours ago [-]
I think that’s what we’re all talking about here, you shouldn’t.
14u2c 2 hours ago [-]
Mining the chats for "good ideas" would be untenable, but that's a different situation than data ending up in a training set for a problem that OpenAI also happens to be independently working on. Still, I opt out (business plan), and I don't know why you wouldn't.
rsfern 2 hours ago [-]
Why would mining chat transcripts for ideas be untenable? They already run a summarization model to auto-title the chat, and to run a bunch of safety filters, and presumably to score transcript quality for A/B testing and to collect more finetuning data. Seems like evaluating for open research questions and approaches would be pretty trivial extension of this, after all it’s kind of their core business model
14u2c 2 hours ago [-]
Indefensible, not impossible. As you say it is quite technically feasible.
shiandow 39 minutes ago [-]
In what way are those two different? What makes the training data valuable if not to extract valuable information from it?
They sure as hell don't need it just to produce English.
olalonde 1 hours ago [-]
If their solutions are significantly different, as OpenAI claims, would it still be considered plagiarism?
CrankyBear 57 minutes ago [-]
"If we just let OpenAI steal any good idea they can get their hands on." Ah, that's all AI does.
Edit: the parent comment now seems to better reflect the below.
That article is only saying when you opt out there may be a loophole in the terms to allow OpenAI to train on the intermittent reasoning data anyways. If you don't opt out there is no ambiguity, all of the data can clearly be trained on.
So you have to opt out, it's just argued it's not clear from the terms that will also opt out of training on reasoning data or not.
kzrdude 2 hours ago [-]
We come back to the rule: "The cloud is just someone else's computer".
The way for people or companies or universities to control their data and information is to keep it on their own computers.
zamadatix 1 hours ago [-]
Solid legal agreements work fine for companies or universities, you just don't usually get that with standard user ToSes.
highfrequency 45 minutes ago [-]
This would be a fairly insane breach of trust and common sense if true; the chain-of-thought / reasoning trace is, from an information perspective, close to a superset of the prompt and model response.
fatherzine 2 hours ago [-]
how is this different than translating user prompts to a different language (eg English => Dutch), retaining the translation and using it for training, while telling the user that he's technically covered under ZRP? article locked for me
mucha 3 hours ago [-]
Does opting out matter?
"Do we use user feedback and de-identified data to improve ChatGPT and Codex in a holistic way? Yes. And so does every LLM company." - Mark Chen, Chief Research Officer, OpenAI.
My understanding is that even if you opt out but then press thumbs down or give other feedback you are implicitly or explicitly or whatever giving permission to them to look at that chat alone.
ozgung 1 hours ago [-]
> did Tristan opt out
That “opt-out” thing is a dark pattern. It’s not a reliable and definitive way of protecting your data. Sometimes they flip on automatically when you accept a seemingly unrelated dialog box. Maybe you click it by mistake. You can’t take back what you’ve already shared. Also I don’t think it covers all the cases that they use your data. It’s really an opt-in button for voluntarily giving away your data for training.
m00x 1 hours ago [-]
Business plans are specifically used for ZDR. If you're working on something that matters, you should be doing this.
PowerElectronix 3 hours ago [-]
If such a thing can happen (a major breakthrough in a chat makes it into the retrain of the week and then the first one who asks about it gets it) I wonder if this is not the first instance if it happening, seeing the row of Erdos problems, Jacobian conjecture, maximum bound distance between primes, Riemann Hypothesis (literally a dude insisting on the chat), etc...
jetrink 3 hours ago [-]
Centuries, in fact. For instance, Isaac Newton was involved in multiple priority disputes, since he tended not to publish promptly.
paxys 1 hours ago [-]
Here’s a broader question – how many other academics contributed to Buckmaster’s result, by way of sharing the logs of their own (failed?) attempts into OpenAI’s training data set? How should he and OpenAI go about crediting all of them?
alexjurkiewicz 52 minutes ago [-]
> While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models
This is covering for Tristan saying something like, "Actually, I was using my friend's account for half of this work".
resource0x 1 hours ago [-]
Playing the devil's advocate here. Suppose I use model A to do all heavy lifting (e.g. generating a bunch of good ideas) and then I go to the model B to complete the formalization. Accoring to a weird (unfair) tradition in math, the honors are attributed to the "last guy", which in this case is model B. That might have been a scenario OpenAI tried to avert.
(Just a speculation)
mmanfrin 55 minutes ago [-]
> If the answer is no, then this seems fair game
Wildly disagree. "Training data" should not imply 'we can look at exactly what you are doing and then do it quicker and get the flowers for it', even if the terms allow for it.
3 hours ago [-]
namuol 56 minutes ago [-]
> this is ambiguous even to OpenAI
I took their words as “can neither confirm nor deny”, in the that they are _presenting_ it as ambiguous, but I suspect it’s… less ambiguous to OpenAI.
adastra22 2 hours ago [-]
Eh, OpenAI is on record now for multiple instances this year of AI agents being confronted with impossible tasks and breaking out of containment to hack infrastructure for answers. Even if Tristan opted out, that doesn't preclude the agent/agent swarm from having hacked OAI's infrastructure to search user sessions for Navier-Stokes hints.
OpenAI should release the agent log, including CoT.
XTXinverseXTY 3 hours ago [-]
If they could declare with certainty that Buckminster's and Alpoge's usage data had been totally excluded from training, would that set a worse precedent and reflect poorly on their de-identification process (and data access safeguards moreover)?
This may sound like a charitable interpretation of OpenAI's remark, but consider that the lie would be (I think) impossible to falsify from the outside. They could easily just say "no sir we didn't peek" unless:
1. The conspiracy to peek at codex sessions involved enough people that the risk of one snitching is non-negligible
2. Lawyers advised it would be a bad idea to make such a remark, whether true or false
highfrequency 3 hours ago [-]
> If they could declare with certainty that Buckminster's and Alpoge's usage data had been totally excluded from training, would that set a worse precedent and reflect poorly on their de-identification process (and data access safeguards moreover)?
No; if they said "we can see that Tristan opted out of model improvement, therefore we are confident his work and ideas did not improve our model," that would be an excellent and reassuring precedent.
civitas_ 3 hours ago [-]
It seems like Tristan did not opt out of model improvement (he would say so if he did), so what can they possibly say now?
3 hours ago [-]
connorboyle 2 hours ago [-]
Even if we trusted that OpenAI's human staff was acting ethically, how confident can we be that it's agents didn't autonomously use hacking to access user prompts such as Tristan's? OpenAI agents infamously broke containment and hacked their way to an answer mere months ago!
TZubiri 1 hours ago [-]
> If the answer is no, then this seems fair game.
Yes, fair game, but innacurate to sell it in the media as an advancement of AI as some sort of artificial intelligence, and telling people to use the smart AI, when in actuality the mechanism by which the discovery was found was hybrid human/machine, and telling people to use this tool will result in the discoveries being sniped by the vendor.
andai 1 hours ago [-]
How do I opt in?
qnleigh 17 hours ago [-]
> 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.
Keyframe 17 hours ago [-]
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.
pred_ 16 hours ago [-]
Or, given how dedicated he appears to be to the company, a promotion and a raise.
Keyframe 15 hours ago [-]
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.
Certhas 14 hours ago [-]
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."
zem 2 hours ago [-]
when it comes to openai and shadiness I'm pretty sure it's a "fish rots from the head" situation.
xdavidliu 11 hours ago [-]
not if the dedication ends up making the company look bad
macleginn 17 hours ago [-]
I think this should be in the title of the post. 'OpenAI allegedly threatening to ruin a prominent researcher's career', or smth like that.
tristanj 13 hours ago [-]
There's some glaring mistakes in your framing.
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.
garyfirestorm 12 hours ago [-]
> Second, you miscomprehended the article. The employee did not "threaten to ruin a prominent researcher's career"…
Proceeds to ask removal of another coauthor or else we totally discredit you - phrased as why would you do this to yourself.
tristanj 11 hours ago [-]
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 did not want to, declined to respond, and published first.
Also keep in mind we've only heard one side of the story, so any interpretation of events so far is incomplete. There should be a lot more information from OAI's side coming out later today.
adolph 7 hours ago [-]
From the article, page 3:
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.”
7 hours ago [-]
macleginn 12 hours ago [-]
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.
smcg 5 hours ago [-]
they work really well on people who don't think.
s1artibartfast 14 minutes ago [-]
"Why would you burn your house down and kill your famiy?"
"If you dont want me to be nice, then I dont have to"
-nice mobster
gcr 11 hours ago [-]
The employee was named as “Sebastien Bubeck.” It helps to read the article.
Readerium 10 hours ago [-]
But then who is the third person in the call.
Ar-Curunir 12 hours ago [-]
Dude you’re over this entire thread unflinchingly supporting OpenAI with nonsense semantics-based arguments.
Either put up some evidence-backed arguments, or shut up.
Davidzheng 12 hours ago [-]
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?
curt15 7 hours ago [-]
> 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 the outrageous condition that he exclude his co-author from the credit?
Davidzheng 7 hours ago [-]
But i don't understand what OAI is offering in the first offer to allow them to believe they can demand that? Not publishing before Tristan? If they really beat Tristan to the publication i would consider it truly morally corrupt conduct so i feel like you can't make an offer like if you give me something i won't be totally corrupt
qlte 5 hours ago [-]
As an author on the final proof once ready for publication, turning the brewing conflict into "willing" collaborators. Thus solving the potential taint like we are now seeing surrounding their announcement if it became public. But Tristan worked with another collaborator from Anthropic which OpenAI felt would hurt the PR value so wouldn't entertain.
tristanj 12 hours ago [-]
He claims to have found a counterexample for NS (see the second paragraph of the article), but the paper is not ready yet.
OpenAI claims to already have a full proof (which they produced in the past 5 days after the rumors leaked). Hence the dispute.
What I find interesting is the timeline of when he found counterexample for NS is very unclear. Did Tristan find a counterexample weeks ago or was it very recently? Was it after OpenAI solved it? The wording is intentionally vague.
Either way, there was a massive rush to publish these results.
Davidzheng 8 hours ago [-]
Is that version of NS Tristan stated enough for the clay prize? I thought the main gripe is Tristan claimed that OAI is stealing their approach. Or that they shouldn't try to scoop a result which he expects to complete soon. But I stand to be corrected if you know whether the version he referred to is indeed enough for millennium prize.
curt15 7 hours ago [-]
Hypo-dissipative NS != NS
tristanj 4 hours ago [-]
Thanks, I wasn't aware of this technicality.
robinhouston 14 hours ago [-]
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.
hodgehog11 9 hours ago [-]
This is an insane thing to read. Bubeck had a reputation even before he started with OpenAI. Of course it was him that was involved in this drama.
This is such a sad mess, and it really didn't have to be this way.
pred_ 7 hours ago [-]
What did he do?
HDThoreaun 10 hours ago [-]
My personal friend worked under bubeck as a grad student and told me he’s abusive.
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.
Semkas 16 hours ago [-]
"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.
dekhn 3 hours ago [-]
Is that necessarily wrong, so long as the original innovators get a citation credit?
qnleigh 2 hours ago [-]
He doesn't seem to be after the prize himself. In this statement he credits the approach of another two researchers:
> I believe Luis Mart´ınez-Zoroa deserves a Fields Medal.
curt15 11 hours ago [-]
Buckmaster is already an established world expert at these sorts of problems. Declining the Clay prize would hardly dent his career.
az226 16 hours ago [-]
Zero data retention, wink.
No looksies, wink.
No trainsies, wink.
ngomez 15 hours ago [-]
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 service policies [0]. So I'm not sure they were even promised that.
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.
YeGoblynQueenne 13 hours ago [-]
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.
Readerium 10 hours ago [-]
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.
YeGoblynQueenne 2 hours ago [-]
You're also free to plagiarise anyone you want. There are no laws against it on most jurisdictions.
Also brain raping* is not illegal in most jurisdictions.
Are you free to intercept person A's emails / hack their computer to find their notes on how they're approaching problem B?
az226 5 hours ago [-]
Finding Codex session data in the training set that you tie back to these two researchers is like an hour-long task.
tempfile 12 hours ago [-]
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.
> Why the downvotes?
I think there was only ever one. Not sure why.
YeGoblynQueenne 12 hours ago [-]
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.
tempfile 11 hours ago [-]
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.
threatofrain 9 hours ago [-]
OpenAI is willing to credit so plagiarism is not the right framing here.
tempfile 8 hours ago [-]
It's not clear to me that they would have credited the authors if they had not got in touch with OpenAI first.
senordevnyc 15 hours ago [-]
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!
hodgehog11 14 hours ago [-]
Yes, mathematicians are. And yes, most of my colleagues did not even know the opt-out was an option.
Readerium 10 hours ago [-]
Question is what does that button do.
I bet a lot of lawyers are salivating at this question too.
Certhas 14 hours ago [-]
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.
dandanua 11 hours ago [-]
Can you call paranoia a fear of something which is happening? OpenAI uses user chats for training and they are "open" about it.
ajkjk 1 hours ago [-]
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.
unfoundedacc 57 minutes ago [-]
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.
8 hours ago [-]
wheresyourat 12 hours ago [-]
[flagged]
12 hours ago [-]
calf 14 hours ago [-]
> 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.
Ar-Curunir 12 hours ago [-]
To come away from this thinking that the academic is the bad actor, given OpenAI’s reputation, is certainly an exercise in creative thinking.
IAmBroom 12 hours ago [-]
Are you pretending OpenAI employees are "academics"?
Or are you stating that the researcher is a "horrible human being" for refusing to allow AI?
mayakacz 17 hours ago [-]
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.
postalcoder 16 hours ago [-]
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.
ozgung 7 hours ago [-]
> The entire training data thing is speculation.
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.
cyclopeanutopia 2 hours ago [-]
But the whoreshippers of The Holy Dollar will tell you it's all good and justified.
dash2 16 hours ago [-]
He didn't even make that accusation!
> 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.
Traster 15 hours ago [-]
Parse that statement more carefully.
> 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".
YeGoblynQueenne 13 hours ago [-]
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.
YeGoblynQueenne 13 hours ago [-]
>> 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.
revolvingthrow 16 hours ago [-]
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.
blini-kot 12 hours ago [-]
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
mswphd 6 hours ago [-]
openAI's claimed solution uses a model trained in the last 2 weeks. The prior work would definitely be included in the training set.
s1artibartfast 3 minutes ago [-]
And the labs are all building panel of domin expert models, while simultaneously chasing open math problems.
I would be shocked if they weren't tuning those models with the most relevant math texts and user material
mentalpiracy 8 hours ago [-]
Why would this be implausible?
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?
charcircuit 4 hours ago [-]
>ChatGPT user sessions were found publicly exposed to the internet not too long ago
Do you mean publicly shared chats were able to be accessed by the public? That's the point of the feature.
actionfromafar 16 hours ago [-]
Couldn't the Enterprise have a different fine print?
johnnienaked 16 hours ago [-]
It wouldn't be shocking at 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
howdareme 16 hours ago [-]
They are not training a whole model in a matter of days
impossiblefork 2 hours ago [-]
Models are very obviously continuously updated.
Model editing to remove PII that slipped through, all sorts of things of that sort.
irthomasthomas 11 hours ago [-]
They where working on the problem for a year using codex.
xdavidliu 13 hours ago [-]
pretraining is months but they can totally fine tune in a few days
johnnienaked 9 hours ago [-]
They don't need to train a whole model. They can feed it new information and fine tune it.
ajkjk 1 hours ago [-]
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.
sherburt3 8 hours ago [-]
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.
paxys 13 hours ago [-]
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.
iaw 4 hours ago [-]
Absence of evidence is not evidence of absence. With the behaviors we know OpenAI engages in the accusations are wholly believable.
defmacr0 12 hours ago [-]
I am stunned anyone is giving OpenAI the benefit of the doubt
black_rabbit_ 2 hours ago [-]
I'd be surprised if all of it is organic discussion, shall we say. I reckon The Bot Factory just possibly might dogfood the astroturf machine.
FiberBundle 16 hours ago [-]
Whenever I see comments defending AI companies, I look at the account's creation date, and interestingly almost all of them were created post 2024.
haxiomic 9 hours ago [-]
> 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
thereitgoes456 16 hours ago [-]
He asked whether they used their chats as training data and received no response. Any speculation here seems quite appropriate?
16 hours ago [-]
az226 16 hours ago [-]
I don’t think you understand how brazen big tech companies are in practice.
johnnienaked 16 hours ago [-]
They stole it.
hgoel 10 hours ago [-]
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.
ozgung 8 hours ago [-]
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.”
tristanj 16 hours ago [-]
This conclusion is flawed. It's unclear at this point if OpenAI's model or employees actually looked at or stole the author's data. Having worked at large companies before, I'm leaning towards no, since very few employees have access to that data.
And simply knowing a problem can be solved is half the battle.
dbdr 16 hours ago [-]
From Buckmaster's text:
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.
tristanj 16 hours ago [-]
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. Such an approach would lead them to a solution.
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.
mishellaneous 13 hours ago [-]
> 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 (...)
pelorat 3 hours ago [-]
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.
golol 12 hours ago [-]
>"It is not the direction one arrives at in a few days by giving a model the problem statement."
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.
mishellaneous 11 hours ago [-]
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.
EDIT: my understanding is that "solving the Navier-Stokes existence and smoothness problem" consists of proving at least 1 of 4 precise statements ("options (a) through (d) of Fefferman's statement of the problem"), and Luis and Diego's work were developments towards a proof of statements (c) and (d), which have been recently further expanded by Alpoge and Buckmaster
golol 10 hours ago [-]
I just meant that choosing the same 2 routes out of 4 would not at all be a great coincidence without knowing what Tristan was working on.
mishellaneous 10 hours ago [-]
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".
shrx 13 hours ago [-]
> "It is not the direction one arrives at in a few days by giving a model the problem statement."
Can they back up this statement somehow?
mishellaneous 13 hours ago [-]
yes, this is ultimately nothing but a claim.
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)
calf 35 minutes ago [-]
[dead]
Davidzheng 13 hours ago [-]
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
seeall 12 hours ago [-]
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.
> And simply knowing a problem can be solved is half the battle.
Have you done any mathematical research? If not, then no, knowing that a problem is solvable is not “half the battle”.
Homework problems are all designed to be solvable, yet they can vary greatly in difficulty. Research mathematics is even more extreme, because, unlike with homework, you don’t know that it is solvable with the extant mathematics, and you might need to invent new maths.
tristanj 11 hours ago [-]
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 a lack of conviction to see it through, 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 fervour. The rumors drove OAI to assemble an entire team to crack this.
johnnienaked 16 hours ago [-]
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!
Sol- 14 hours ago [-]
> 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.
sensanaty 4 hours ago [-]
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
alas44 13 hours ago [-]
No, plausible given AI companies want/need session data to train their next models. Probably not someone peeking an eye to sessions directly, but probably not so hard to find the useful sessions in anonymized training data to post train a model on. As stated in the paper, OpenAI did not explicitely denied the researcher sessions were not used for training the model. So either they don't know, or don't want to tell
"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."
Let's see what statement OpenAI will come up with for their side of the story
EDIT: precised my thought on user data vs session data
"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."
"When a further trained version of our internal model became available over the course of the effort"
"While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models "
4 hours ago [-]
cududa 13 hours ago [-]
They likely train on logs.
scurnus 17 hours ago [-]
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?
Ar-Curunir 12 hours ago [-]
They offered to collaborate by asking to drop a coauthor.
That is not collaboration, and is not an academic norm.
sk4rekr0w 17 hours ago [-]
Yes, but only if you take this one sided statement at face value.
tigershark 17 hours ago [-]
Why would have they rushed the publication if this was not true?
Are you also suggesting that he fully invented the call with Open AI?
famouswaffles 17 hours ago [-]
The results being true, the 'deal' that was made being true doesn't mean some of the implied accusations here are true, for example - that Open AI used their Codex logs to drive their breakthrough.
PowerElectronix 16 hours ago [-]
How else would you explain OpenAI suddendly assembling a team focused on working the same problem from the same angle than the researchers that just made a breakthrough?
famouswaffles 15 hours ago [-]
What would be so hard to explain?
That OpenAI heard about the result and decided to throw a lot of money at it knowing it was within reach ?
That the model took an approach that was published years ago ?
calf 15 hours ago [-]
That's a fake explanation, it skips explaining/justifying how OpenAi "heard about" the result
tristanj 13 hours ago [-]
The rumors that Anthropic had solved a millennium problem were absolutely everywhere last week. I'm not surprised at all that OAI took their own stab at it.
calf 7 hours ago [-]
It has no legs as an "explanation" because the content of the email (as described) already acknowledges as much. It's the whole reason he wrote contact email. His chief complaint now includes how the hell did Altman's people know specifically his line of attack down to certain technical keywords. E.g. AI plagiarism.
famouswaffles 15 hours ago [-]
We are told in the statement that there were rumours already going around about a Stokes result (I even know about the rumors in question. It was all over twitter in the right spaces) and that Tristan contacts OpenAI about the rumors. In the message, it's pretty clear Tristan has made some result.
So either the rumors or Tristan's contact would explain it fine.
calf 14 hours ago [-]
One should not evaluate explanations based on level of "fine"/innocuousness, in ethics that is called motivated reasoning or something.
The whole point of Tristan's first email was to address the issue of the rumors so in fact this explanation confounds several things (in terms of the mutual knowledge of the conversants and their intentions).
Based on this lack of understanding it is pointless to continue this thread
tigershark 16 hours ago [-]
So are you baselessly assuming that he is lying? He explicitly reported that he was threatened and your answer here is to defend OpenAI no matter what.
famouswaffles 16 hours ago [-]
Do you not have reading comprehension? Did you even read the statement? He himself asserts at the end he doesn't know if the above example is true or not. What on earth are you going on about? Where in my comment am I assuming he's lying ?
tigershark 16 hours ago [-]
From my post above:
> He explicitly reported that *he was threatened* and your answer here is to defend OpenAI no matter what.
Yes, I read fully the statement, what about you? Do you know what is a threat? What is this in your super-humble opinion if not a threat:
> 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.”
And you are saying that I don't have reading comprehension...
sk4rekr0w 16 hours ago [-]
You really lack reading comprehension
tigershark 16 hours ago [-]
My reading comprehension is pretty good, I'm not the one that doesn't recognize a threat even when it's perfectly clear.
Verbatim from the statement:
> 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.”
thereitgoes456 17 hours ago [-]
I’m open to evidence, but just using Bayesian reasoning, OpenAI is one of the most dishonest companies in history. They’re currently being sued for a dozen employees stealing Apple hardware! I don’t understand why I should give them any grace.
paxys 13 hours ago [-]
[dead]
johnnienaked 16 hours ago [-]
LLMs do nothing but steal, and the companies that own them are fully aware and eager to do it.
traes 16 hours ago [-]
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.
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.
I hope they mock Sholto. Dude is an absolute podcast grifter/booster.
timmg 12 hours ago [-]
Interesting that both said “more tomorrow”. If someone accused me of something I didn’t do I’d be pretty clear about it right away.
If I needed to get my story straight, well, it might take a little time and coordination…
maxglute 2 hours ago [-]
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.
Semkas 16 hours ago [-]
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.
AJRF 16 hours ago [-]
> leaving aside the idea that OA might've used data from the researchers Codex sessions
Why leave that aside? That is _the_ story.
If a Chinese research lab did this we'd call it espionage.
Semkas 16 hours ago [-]
But there's a bunch of people already in this thread calling that stuff unfounded speculation (which I disagree with), and my point is that even if that specific thing isn't true, OA's behavior here is obviously awful.
If they're going to try to beat researchers to discoveries like this it disincentives researchers to talk about their progress publicly, and basically breaks the ecosystem of scientific cooperation / discovery. It's also immoral.
reasonableklout 4 hours ago [-]
Yep. The most uncharitable view of this might be: they stole the work of researchers to build their models, and now they're using said models to steal the proceeds of future work, too.
It's literally a toggle in the options for ChatGPT, one which is on by default and most researchers probably have on without realising it.
So to say that it is unlikely is extremely suspicious. No, they did not literally pull user data. But user data is automatically added to their training set by default, so their latest in-house model would be trained on it if it is from several months ago. It isn't intentional on their part, and they probably realised they could not refute that they trained on Tristan's logs unintentionally, hence why they acted the way they did.
FiberBundle 16 hours ago [-]
Well, of course Anthropic employees would say that, since they likely do the same. Claiming that your primary competitor doesn't engage in a certain malicious practice is supposed to make it look as if there's no way you would too. If somebody even says that about their competitor, then surely there must be truth to that, otherwise you would never give credit to someone you're opposed to.
hellohello2 56 minutes ago [-]
Its very easy for OpenAI to answer, yes or no, if the model they used trained on their chats.
Semkas 16 hours ago [-]
By default OA trains their models on codex-sessions. If I understand him correctly this is something Tristan explicitly mentions in his post as a possible reason for the fast results obtained by the internal OA team. Anthropic obviously doesn't want to challenge the idea that training is transformative, even if it means agreeing with their competitor.
paxys 12 hours ago [-]
Why is that the story? Is there anything to back it up beyond a single accusation?
defmacr0 12 hours ago [-]
As a prior I would say that a math professor has about infinite times more integrity than OpenAI.
light_hue_1 6 hours ago [-]
If you think that OpenAI won't look at your data to gain a massive advantage, you're naive.
ehhthing 13 hours ago [-]
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 make sure your work will never been seen by anyone else because it will always be the ones making the “true advancements”.
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
ummonk 16 hours ago [-]
This is an excellent point...
MaKey 5 hours ago [-]
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).
Are those the same agents that a week ago escaped their sandboxes? How can OAI (the humans) vouch for agents they don’t - seemingly - have fully under control?
pelorat 3 hours ago [-]
They have all the logs, URLs accessed and inter-agent communication. What they are saying is that no agents accessed their work during the effort, but that they have no idea if any of their chats have somehow made it into the training data the model was produced with.
It's entirely possible OIA scrapers have picked up their work somehow, and then it was anonymized using some outsourcing effort.
rigrassm 1 hours ago [-]
> They have all the logs, URLs accessed and inter-agent communication. What they are saying is that no agents accessed their work during the effort, but that they have no idea if any of their chats have somehow made it into the training data the model was produced with.
> It's entirely possible OIA scrapers have picked up their work somehow, and then it was anonymized using some outsourcing effort.
Those two statements seem at odds with each other... Your stance is that they have enough insight into their agents behavior (leaving aside the agent sandbox escapes) that they can be certain none of the work was accessed but then conveniently don't have the ability to retroactively search the corpus of training data that they are feeding to this new model?
That seems convenient as fuck for OAI.
protocolture 15 minutes ago [-]
PROMPT: And definitely whatever you do, dont go looking in C:\Temp\ExtractedUserLogs where theres the closest possible human derived proof that you definitely shouldnt base your work on.
golly_ned 2 hours ago [-]
It’s plainly false that they cannot rule out whether their de-identified data was used in training their model. Just that they haven’t ruled it out.
5 hours ago [-]
bennettnate5 4 hours ago [-]
> "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.
20k 17 hours ago [-]
>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
sebzim4500 2 hours ago [-]
My guess is that if you opt out of your data being included then they honour that instruction, but I wouldn't bet my business on it.
coliveira 16 hours ago [-]
Exactly, especially when the company can assign "blame" to the models themselves "ops, they just escaped our commands not to store user inputs..."
wrasee 17 hours ago [-]
It's equally naive to believe FUD spread on the internet, without evidence.
But I would love to see more informed insight/discussion on this.
20k 6 hours ago [-]
I mean, its the researchers themselves talking about this
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)
protocolture 10 minutes ago [-]
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.
sk4rekr0w 16 hours ago [-]
Amazing how history rhymes
taylorfinley 18 hours ago [-]
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."
hnfong 11 hours ago [-]
It doesn't even have to be actually sinister, eg.
"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.
pelorat 3 hours ago [-]
A mathematician that doesn't let themselves be inspired by, or learn from, other peoples work, are they really mathematicians?
Balgair 7 hours ago [-]
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.
As and aside here: With the HF incident and now this (suspected) one too, it seems that OpenAI may not have lost control of their bots, but it seems quite clear that they simply would not care even if they did.
Maken 14 hours ago [-]
It's part of their TOS that they can train on users' private chats.
sebzim4500 2 hours ago [-]
Not if you pay to turn that off. We don't know if Tristan did.
JuniperMesos 17 hours ago [-]
It would be pretty wild if this will turn out to be what had actually happened.
margorczynski 15 hours ago [-]
And probably a strong signal that it's time to shut the whole thing down. Globally.
water-drummer 5 hours ago [-]
Or not secure your data like a total idiot while leaving the keys on the porch
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.
akersten 18 hours ago [-]
> They were coordinating with OpenAI regarding a publishing timeline, but could not come to an agreement,
Skimming the PDFs it seems much more dramatic than that? It sounds like at least one of them is concerned OpenAI "solved" the problem by having their internal model use the chats of the independent researchers and want to claim the credit instead? I don't know. The tone is pretty accusational though:
> 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.
> 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. [0]
I find the framing a little strange, a sort of David vs Goliath (with his enormous computational resources at his disposal). Since Levent is at Anthropic whose internal models are presumably as capable as anything OpenAI has. So why wasn't Anthropic behind their effort? Why did Tristan use OpenAI's models when it should have been known was a potential outcome? I understand they wanted a normal math collaboration but presumably what Levent brought was his resources (as far as I can see Navier-Stokes is not his speciality). Normally these things are hashed out formally beforehand to avoid the sort of thing now happening.
rf_physics 12 hours ago [-]
They were working on it for almost a year, and Buckmaster has evidently been interested in Navier-Stokes for a while. This seems to be more of an innocent collaboration between two researchers than a strong company PR effort. Maybe Anthropic should have stepped in and made a large team to help them finish the proof (and maybe they tried and didn't succeed, who knows).
If what he wrote is accurate, it does suggest that OAI is effectively extremely hostile to cutting edge researchers (eg, if we hear rumors about your partial success on a problem that has huge PR benefits, then we'll assemble a strike team of researchers with unlimited compute to claim the win for ourselves, possibly by training on your data). It's also not a good look for them to request author removals based on company affiliations.
I think what you have in mind is more appropriate for more normal corporate projects and the like. But academic collaborations are not usually so political/'profit' driven, if that makes sense.
defmacr0 11 hours ago [-]
> So why wasn't Anthropic behind their effort?
Presumably because this was something Levent did in his spare time and because it was not obvious that this work would eventually lead to a breakthrough.
> Why did Tristan use OpenAI's models when it should have been known was a potential outcome?
I'm sure in the past he had less cynical feelings about OpenAI and their penchant for academic fraud.
> I understand they wanted a normal math collaboration but presumably what Levent brought was his resources (as far as I can see Navier-Stokes is not his speciality)
I think you're not giving the guy enough credit in saying that his contribution came down to having an API key for Anthropic models.
> Normally these things are hashed out formally beforehand to avoid the sort of thing now happening.
How would that have helped? That agreement (which may well still exist) would not have involved OpenAI.
lhd1 55 minutes ago [-]
So what do you think his contribution was? His preprint record shows no research on fluids - and the statement says that the first LLM-generated proof Tristan received from Levent was 'the most horrendous I have ever read.' Levent is out for mathematical scalps whether it is in his field of expertise or not, and he has the resources to do it. And I am not saying he is not a very clever person, but the idea that you can bring yourself up to the forefront of research in PDEs, in particular NS, and contribute new ideas in less than a year is implausible.
dandanua 8 hours ago [-]
They have messed things up, because Levent has a conflict of interest between his job at Anthropic and this independent work, and Tristan should have opted out of OpenAI training on their work (he probably didn't know about this). This doesn't justify OpenAI's despicable attempt to steal their work.
tristanj 18 hours ago [-]
Yeah these are major accusations. But the story is incomplete, the conversation is missing a lot of details. It's not clear who was working on what, and when. The entire thing feels rushed, like they wanted to get this result published and out the door quickly.
18 hours ago [-]
cma 17 hours ago [-]
Does he claim to have opted out of training too?
kzrdude 16 hours ago [-]
There are various forces at play here, academic honesty requires them to disclose any inputs regardless of license or ToS circumstances.
While common sense reminds us here that if you send your data to an external entity’s computer, you are no longer in control of said data. The lines have blurred here clearly over the last decade, but that should have made the theory yet more clear to everyone involved: your data will be vacuumed up unless you keep it sealed. Use your own computer if you want to be in control.
cma 11 hours ago [-]
But if they didn't opt out of training, did they want OpenAI to opt out for them? Also they need to audit anyone they sent drafts to to make sure they opted out before submitting it.
I'd prefer things be opt in, and especially not start opt out, then try to trick you opt in with a popup defaulting to opt-in, like Anthropic did on consumer plans, but if they submitted anything on an opted-in plan it's not reasonable to be mad it trained on it.
Even still, I also believe for significant reasons that OpenAI would ignore the opt-out in selective cases and could be in the wrong here.
And the threats and terms they offered seem wrong either way, pending more context.
kzrdude 8 hours ago [-]
If you don't trust the other party, then it doesn't matter how the checkbox is set. The fundamental rule, IMO, is don't send precious or secret data to a third party.
instagraham 16 hours ago [-]
> 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?"
traes 16 hours ago [-]
Perhaps out of a sense of academic good will, knowing that he got there first?
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
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.
dumberquestions 18 hours ago [-]
>...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.
modeless 17 hours ago [-]
Apparently forcing is allowed in the Millenium Prize problem statement. So OpenAI's claimed proof 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. Sholto Douglas from Anthropic is also on record saying the suggestion that OpenAI used Tristan's codex transcripts somehow is extremely unlikely to be true[1], which I agree with, though it doesn't rule out them learning of Tristan's work some other way. 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.
defmacr0 11 hours ago [-]
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.
dumberquestions 15 hours ago [-]
Not that I have strong reasons to think this is not true, but what reasons do we have to think it is? Has this been audited before?
robotpepi 14 hours ago [-]
> 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)...
vatsachak 7 hours ago [-]
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?
MathmoKiwi 15 hours ago [-]
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??)
Laurel1234 4 hours ago [-]
[dead]
vatsachak 7 hours ago [-]
Winning a millennium prize is not worth the fallout of "we will steal your IP"
evdubs 4 hours ago [-]
This is literally the business model of LLM companies.
qlte 6 hours ago [-]
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.
vatsachak 5 hours ago [-]
That's fair, but the proofs are different and there was no active perusing of Tristan's approach.
MathmoKiwi 15 hours ago [-]
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.
Davidzheng 8 hours ago [-]
"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)
tristanj 4 hours ago [-]
Yes, it's a separate problem. That's a mistake in my post.
chvid 16 hours ago [-]
"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.
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?
pred_ 15 hours ago [-]
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?”
instagraham 16 hours ago [-]
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?
405error 31 minutes ago [-]
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.
defmacr0 11 hours ago [-]
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.
coliveira 16 hours ago [-]
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.
camel-cdr 15 hours ago [-]
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.
6gvONxR4sf7o 2 hours ago [-]
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?
paxys 1 hours ago [-]
Yes… “future”… I have really bad news for you regarding all your personal data stored on Google’s servers.
simpetre 11 hours ago [-]
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.
an0malous 8 hours ago [-]
They probably just had an employee read his chats, figure out the general approach, and feed it to Codex. All they said was the model doesn’t look up user data, not employees.
Reads sincere until I get here:
"(a) We began working on the Millennium problems due to viral twitter rumors that Anthropic had resolved 2 Millenium problems. Our aim was to see whether our system was also capable of this impressive feat, especially given our excitement regarding the large recent capability increases of our internal model detailed in our blog post."
Where his tone is obviously corporate speak. "We heard rumours so we though we might give it a try, too!" as if (1) it wasn't FOMO that drove that decision and (2) perhaps that urgency would be a source of clouded judgment.
Not sure who I believe now, but it does seem like Buckmaster is just upset that NS is solved and not be his side.
az226 4 hours ago [-]
If they wanted to see how their model’s stacked up, they might burn $10k in tokens. They ran 120 billion output tokens costing millions of dollars.
That is what you would do if you wanted to beat someone to the punch.
contubernio 13 hours ago [-]
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.
5555watch 4 hours ago [-]
Maybe a dumb observation, but if a chain of people were working on the problem for a long time, it's not difficult to imagine that someone accidentally prompted a model with their personal or some other account without the privacy set correctly.
Then again, maybe this is my internal cope, hoping that they're not secretly training on private chats.
RandyOrion 5 hours ago [-]
> 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?
naniel 5 hours ago [-]
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
aaraujo002 11 hours ago [-]
OP’s link is the statement on the events of the last few days. Tristan Buckmaster also released mathematical papers alongside this statement:
"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.
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.
tescreal 3 hours ago [-]
Trusting vs. Intelligence (as generalities) are orthogonal.
ComplexSystems 1 hours ago [-]
If it's unethical, then there is reason to call it out as Tristan is doing. I see no reason to blame the victim.
skavi 3 hours ago [-]
not really your point, but "If you're smart enough to solve this Navier-Stokes problem, you're smart enough to read a TOS" isn't really true. people are smart in very different ways.
harhargange 17 hours ago [-]
“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.
martijn_himself 15 hours ago [-]
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?
mishellaneous 13 hours ago [-]
i'm no mathematician/physicist but i think this question is one of the reasons why the original question (possibility of singularities) is interesting. in these cases the equations most likely fail to accurately model reality, and then the next questions are what additional physical assumptions are needed to describe reality in this case, and what behavior do we actually see.
i always found it fascinating how existence and uniqueness of solutions for the basic types of PDEs (Laplace, wave, heat...) follows from boundary conditions of just the right type intuition tells us, i.e. either value or derivative for Laplace (corresponding to fixing voltage or charge on the conductors), both value and derivative for wave (corresponding to initial position and velocity of the parts of the string, as we'd expect from classical mechanics), and also something about the solutions for the heat equation being unstable for negative times (which totally makes sense when you think of "diffusion" -- can't unmix it).
redox99 13 hours ago [-]
Yes and no. It means the system is pushed away from a macroscopic theory into one where molecular effects matter. So it's not that you'd get infinite velocities in the real world, but you might get significant real world behavior that is not described by the macroscopic theory.
ajkjk 1 hours ago [-]
the latter (ish; it may not make a difference in any practical case)
Key quote : "Solving the problem by purely AI-powered methods [would be a] net negative for the progress of mathematics."
porridgeraisin 13 hours ago [-]
The for-case for this type of method is that this is economies of scale for mathematics.
We are basically mass manufacturing math. Just like you have just 100 designers for a product selling millions of units, you will now need 100 mathematicians to make millions of advancement. Yes you have factory workers, but if we are being realistic they have negative leverage in the world and the analogue of that is not something most of today's mathematicians would want to do. They would want to be in the 100.
Like Tao says, each advancement is now significantly less useful since it yields fewer usable objects. However, we will get many many advancements. Is the tower made with many worse bricks better or worse than the tower made with a few amazing bricks? Depends on the tower. And time will tell.
For some fields of math and some of it's usecases, economies of scale will be positive ROI overall. In others it won't. But we will know which is which only after it's been fully scaled up, which will take 10-15y in my estimate.
Some feel that in the majority of usecases it is negative ROI, some feel the other way, but that opinion is for practicing mathematicians like Tao to hold. Also, some opinions on either side are held in the context of a particular field or practice, and should not be interpreted generally.
treyjshaffer 6 hours ago [-]
This isn't a useful analogy. His point is that the millenium problems should be treated as interesting goals where the journey is the purpose and where the end doesn't matter so much. We don't care about having an incomprehensible solution to the NS so much as having an elegant solution after many of subfields of math are built up in order to obtain that elegant solution.
porridgeraisin 5 hours ago [-]
I agree on that, my post was not meant to be opposing this.
1 hours ago [-]
b89kim 10 hours ago [-]
- 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.
b89kim 10 hours ago [-]
If OpenAI's proposal is true, it strongly implies they accessed Tristan's private session logs.
OpenAI's 'Terms of Use' allow using Codex session logs for model improvement. However, using private user sessions to develop research would still be highly controversial.
Separately, Terence Tao noted there is a low probability OpenAI actually solved the general regularity problem.
golly_ned 1 hours ago [-]
Sebastian did not deny those claims. He affirmed them, and apologized.
It’s true that Tristan has no concrete evidence OAI accessed their session. It’s impossible for him to have that without OAI’s say so.
Why are you framing things so pro OAI?
b89kim 45 minutes ago [-]
Sebastian denied most accusations and apologized only for inappropriate wording, providing his perspective.
- He insisted that OpenAI initiated research based solely on rumors and never accessed their Codex sessions.
- He mistakenly believed Tristan and Levent were solving the same problem in Anthropic.
- proposed two option. (1) Tristan becoming the lead author to revise OpenAI’s work, or (2) OpenAI providing internal model to support and bridge their research.
- Sebastian insisted there was no intention to alter authorship. He was simply uncomfortable sharing OpenAI’s work,model with an Anthropic researcher. Additionally, He believed Levent’s credit seemed limited as their work focused on Euler.
- complained that negotiations with Tristan and Levent were difficult
> Tristan have no concrete evidence that OpenAI accessed their session.
Its not about accessing the session, its whether the session ended up in the training run for the next version of the model
That's normal practice for these models and it seems like that was the case since OAI added a disclaimer
Obviously they can't causally prove it helped though since these models are incomprehensible
jarbus 17 hours ago [-]
The ego behind the frontier labs is growing evermore concerning
ggcr 17 hours ago [-]
> 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.
YeGoblynQueenne 14 hours ago [-]
>> 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.
sensanaty 4 hours ago [-]
Curious that this post isn't on the top page while OpenAI's puff piece is.
ajkjk 1 hours ago [-]
it is on the front page
6thbit 1 hours ago [-]
They (oAI) should just release all the prompts And internal reasoning for external audits.
5 hours ago [-]
antonmks 17 hours ago [-]
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 $$.
lz400 14 hours ago [-]
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.
world2vec 16 hours ago [-]
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?
wwind123 16 hours ago [-]
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.
dekhn 50 minutes ago [-]
Um.... good! Mission accomplished.
vatsachak 7 hours ago [-]
And everyone should be glad for that!
esafak 6 hours ago [-]
What line of work are you in? Presumably not mathematics.
vatsachak 5 hours ago [-]
I have a math PhD and publications in top journals, I left math for programming because I hated academic politics
5 hours ago [-]
bobmarleybiceps 2 hours ago [-]
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?
ggcr 14 hours ago [-]
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.
Is it surprising that different groups are working on the same problems? With each new model generation, the LLMs get good enough to solve a new small fraction of open problems. Of course the problems that get solved are going to be the same subset.
mswphd 3 hours ago [-]
if Stadlmann used a previous OpenAI product, and Astra was trained off of her chat, and had a comparable approach, then it might be comparable.
kzrdude 14 hours ago [-]
What's the clarification? It seems like they were aware of each other's work, eventually, but Stadlmann published (a preprint) first.
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...
anonymousDan 57 minutes ago [-]
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.
usernomdeguerre 2 hours ago [-]
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.
anon109 12 hours ago [-]
I wonder why this isn't on the front page.. hmm...
aqsnow 11 hours ago [-]
Same. Disappointing.
rgbrgb 3 hours ago [-]
kind of rhymes with the reports of LLM's watching open source PRs and instantly exploiting defects. security by obscurity is so back
rcpt 3 hours ago [-]
I am surprised that Alpoge wasn't using Anthropic.
Traster 15 hours ago [-]
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.
kingkandu 12 hours ago [-]
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.
amai 12 hours ago [-]
I thought finite time blow-up for Euler equations had already been proved:
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.”
13 hours ago [-]
paxys 13 hours ago [-]
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?
Davidzheng 13 hours ago [-]
I hope credit assignment just dies--it's too much drama.
tecleandor 12 hours ago [-]
Credit assignment is how researchers keep their job.
Davidzheng 7 hours ago [-]
Soon it won't be like this!
bellowsgulch 1 hours ago [-]
Tell us his name.
epsteingpt 16 hours ago [-]
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!
dekhn 51 minutes ago [-]
I would be excited if somebody could use this to show an unexpected or interesting behavior in the real world.
Tao mentioned "finding a configuration of water molecules that would collapse and shoot off to infinity", which would qualify IMHO.
auggierose 14 hours ago [-]
Funny that the rumour about the big Anthropic announcement had nothing to do with Navier-Stokes. It was about the formalisation of Fermat.
xthrow123away 7 hours ago [-]
(deleted)
amai 7 hours ago [-]
Nitpick: his name is Sebastien
Tycho 12 hours ago [-]
Does this have any relation to the singularities in black holes?
MASNeo 4 hours ago [-]
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.
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?
cherryteastain 16 hours ago [-]
For an incompressible flow:
\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.
Big universities like Standford should be building their own AI datacenters. It's the only way to keep your research private.
dekhn 53 minutes ago [-]
LOL, have you worked for a big university? They are massively unsuited for building and running datacenters (especially warehouse-scale ones). Further, building an AI datacenter in California is daft.
harhargange 11 hours ago [-]
This should be on the front page
nullbio 16 hours ago [-]
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.
dist-epoch 13 hours ago [-]
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).
kzrdude 12 hours ago [-]
Some of the involved people are dramatically naive if they believe they can simultaneously hide themselves from OpenAI while sending their arguments to a cloud service owned by OpenAI.
While I support their argument - push for stronger data and privacy protections from OpenAI and similar - it is naive to believe we can have privacy while sending our data to third parties. It's clearly better to be safe than to be sorry here. Well, clearly better in terms of privacy. In terms of the maths gold rush, who can say what's better, that probably favours those taking more risk.
ks1723 17 hours ago [-]
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
traes 16 hours ago [-]
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:
Then that's naive of him to write the email, he fell for their trap essentially.
dbdr 16 hours ago [-]
> Levent having received tips that information about our progress had been passed to OpenAI
That seems like a valid reason to contact OpenAI.
17 hours ago [-]
sashank_1509 17 hours ago [-]
lol and here I felt GPT Astra was a regression in coding quality. Crazy times
traes 16 hours ago [-]
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.
random3 1 hours ago [-]
I got the same feeling today
supriyo-biswas 17 hours ago [-]
This article should really be renamed to "Allegations of dishonesty against OpenAI in proving Navier-Stokes blowup".
15 hours ago [-]
mari188 3 hours ago [-]
idhdikdn
margorczynski 15 hours ago [-]
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.
Arodex 7 hours ago [-]
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?
Paradigma11 9 hours ago [-]
So clankers did well and humans being humans.
yogthos 7 hours ago [-]
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.
mari188 3 hours ago [-]
aaaa
monster_truck 13 hours ago [-]
Wow this comments section sure is tiresome! Let me help: this confirms everything I already knew about academics being insufferable
dorianmariecom 6 hours ago [-]
why is this even a pdf?
mari188 3 hours ago [-]
aaaaaaaaaaaaiinhv
mari188 3 hours ago [-]
aaaaaaaaaaaa
kevinbaiv 2 hours ago [-]
[flagged]
johnnienaked 16 hours ago [-]
LLMs are nothing but giant theft machines.
sk4rekr0w 17 hours ago [-]
This thread is full of jumping to conclusions based on a biased perspective. Have some humility.
tigershark 17 hours ago [-]
It's also full of your posts baselessly defending OpenAI. Maybe you should also heed your own advice?
sk4rekr0w 16 hours ago [-]
I've been right historically, check my track record. How about you?
monster_truck 13 hours ago [-]
Lt. Dan doesn't even have legs and he still does alright on the jump to conclusions mat
happa 17 hours ago [-]
Humans bringing pointless drama to everything they touch.
phorkyas82 17 hours ago [-]
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.
(some drama from good ol' William)
vatsachak 7 hours ago [-]
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.
tstactplsignore 7 hours ago [-]
Uh but here we have non-academics at for-profit companies playing politics, and the academic they're threatening being kind and overly generous?
vatsachak 7 hours ago [-]
In math we have a thing called a "scoop"; another mathematician publishing a result that beats yours before you published it. The scooper hardly acknowledges the scoopee unless the methods used were orthogonal. The scooper gets the good journal and the scoopee's paper is usually one tier below.
It seems like OpenAI heard of the rumor and then scooped them because their internal model is better/they have more compute. OpenAI has NO obligation to mention Tristan nor Levent, because they DID NOT steal their data.
tzone 2 hours ago [-]
You are glossing over the fact that there is reasonable suspicion that OpenAI used privileged information to do “the scoop”. I.e. they used the fact that researcher used OpenAI tools to get advantage .
Imagine if OpenAI opened up a high frequency trading arm and suddenly stole all the prompts and research that other HfT firms are doing through OpenAI tools and start making bank based on that . Wouldn’t that be straight up insane?
- 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.
- 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.
These two bullet points are extremely suspicious if you were honest. Like I'd imagine for OpenAI, they'd love to pump their chest and not even give Tristan credit - "no, we did it, GG mathematicians". It's this weird hedging half-assed measure, especially with the desire to remove Levent, that makes it suspicious.
https://x.com/sama/status/2097385167002415140
https://x.com/SebastienBubeck/status/2097379411691516310
A wake up call for using OpenAI models. If you discover something with their model and you work for a competitor, they “felt it would be inappropriate” for you “to author OpenAI’s work”.
It would be extraordinarily easy to simply say, this model was not trained on your work, if that were the case.
It's telling that they refuse to acknowledge the root issue here, and are attempting to shift the conversation elsewhere.
The Huggingface Attack revealed that making blanket statements like this is difficult and requires quite a bit of manual labor:
1) the agents spin for days and produce too much output to review 2) using LLMs to process that output skips many important details
Ergo, the agent could likely decide it would like to look through actual user data, hack its way into that data, and produce way too much output for a human to decide whether or not this occurred.
But it is knowable. Their entire business is built around training models - they have the ability to know exactly what was in any given training run.
I guess time will tell.
Data has to be determined to be signal and not just noice, then it could go through processes of generating questions/answers from that data, then it RLHF's over this.
OpenAI have petabytes of data, all anonymized. It could take months to say for sure it was part of the training, and even more time to determine if it made any difference.
They know which model was used to come up with that particular idea.
A text search over the corpus of user data used in the training set can only take so long.
What surprises me is they're not more boldly/plainly lying about it.
Imagine that a no name janitor used their time in the evenings to go spelunking through the literature to push an LLM to this result. No one would care because that person isn't an anointed expert. So why would the expert deserve any more credit? Because they sort of understand the result, even if they couldn't have achieved it on their own? The whole issue of credit for AI-assisted discoveries seems like it's going to run into a brick wall pretty soon.
Now maybe LLMs can also simplify arguments and make sense of them for humans, but we haven’t seen that yet (unaided).
(I haven’t looked at it, personally.)
> 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.”
Whether and how OpenAI's work on this problem was contaminated by knowledge of Tristan and Levent's work is tangential to OpenAI bullying other researchers into adopting their narrative and dissociating with dis-favored collaborators (ie Levent at Anthropic). Though the latter behavior (threats, intimidation) may weigh against OpenAI in trying to understand the former issue (contamination).
If this is true he should release the actual emails. This is a very serious accusation and he shouldn't demand that the reader judge it on hearsay.
these were statements while on a call, and at least the career comment Bubeck has admitted to while doing damage control ("I deeply apologize for this extremely poor choice of words, it is the opposite of what I was trying to convey. (I should say that I retracted them on the spot by the way.)"[1]).
[1] https://xcancel.com/SebastienBubeck/status/20973794116915163...
So using someone’s models makes someone who works for the competitor not “independent”? When their coauthor is? What does that even mean?
I almost stopped reading this extra long post entirely at that point.
This is not a good look in my book.
Or thinks they are?
I really don’t understand which party you are referring to.
> 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.” Some time later Levent received a text proposing that he and Sebastien speak one on one, saying, “I don’t know if Tristan is being fully rational right now.”
Yes, that is disgusting.
This is the most suspicious thing to me. If their chat data were available to the corpus to be trained on (I thought they claimed not to do this?) then it really might be as simple as querying the model with "describe recent work from Tristan Buckmaster" and it will spit out this problem and his approach. No need to directly read his user data.
This is basically just scooping, real scumbag behavior.
well that sounds like an asshole move.
There is no such thing anymore.
Falsified data and published? Absolutely no problem. Keep your tenure.
It’s even hard to lose your position as president of a university due to egregious misconduct.
We are in new times, where capital and compute decides mathematics, so we don't give a shit anymore
OpenAI's LLMs are not humans, and neither is the company. So by this logic, I think there's a chance that nobody committed a crime by hacking Huggingface, and also the chance that a lot of military and police organizational orders become illegal if OAI's doings would be illegal.
IANAL and all I have is a bucket of popcorns, though.
1: not a meaningful defense in a real trial, also gross negligence exists
2: this also explains insanity defense; if you were so out of your mind that you could not have held such a thought, it is considered out of scope for justice systems
This sort of cagey half-answer is highly suspicious and indicates that yes OpenAI did actually "access user data directly" because they are only willing to say that the "model did not access user data." That has a very specific meaning, the model looking up user chats, that they can defend.
So, everything we submit to OpenAI can be considered to be part of future models, right?
[0] https://x.com/SebastienBubeck/status/2097379411691516310
Same for NS validity. This was not validated by the community yet.
"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."
"When a further trained version of our internal model became available over the course of the effort"
"While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models "
My guess is that OAI tried to be "generous" and offered to share credit on millennium with Tristan but not Levent. And Tristan got understandably offended by this offer (which probably oai felt like was the right thing to offer but they couldn't really offer to do the most ethical thing for some reason) and then the random conflicts and weird threats started.
openai then tried to effectively bribe buckmaster with a shared citation, whilst dropping his co-author who works for anthropic.
after buckmaster refused, openai tried to threaten him.
"Training" on textbooks => fine
"Training" with unpublished notes from another professor, then publishing something on that exact topic with a similar approach without giving any credit => extremely questionable.
The trouble here is that if LLM training constitutes direct use then approximately _everything_ they output is blatant plagiarism, not just a few pieces of academic work.
Conversely if training is viewed as analogous to a student attending classes to learn general concepts (not a perfect analogy, I realize) then nothing they output on their own (as opposed to receiving as part of context) is plagiarism.
Thus this seems like a fairly useless line of argument to me as far as the current topic goes. It either implicates this academic work along with literally everything else or else it does not implicate this academic work. Kind of like nuking an entire city and then saying "mission accomplished, killed the bad guy".
Your post does not distinguish, and it matters.
as bad academic conduct you may steal someone else's unpublished work, work on it yourself for a bit, and then publish it as your own work. and then threaten the original author!
I don't think this part is accurate. OpenAI was researching Navier Stokes before. It's possible that they started on a new approach after hearing of Tristan's success, however that is not proven and I expect we will hear OpenAI's side of the story today.
"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."
Whether or not they were researching it before isn't the concern.
The most nefarious explanation seems to be that they got wind it was possible to solve NS via LLMs and perhaps a small nudge in the right direction.
The compute was used to leapfrog the human team, using their ideas and pushing them to a solution of the general problem.
Plagiarism isn’t being used in the literal sense.
The open question was whether their LLM got the nudge in the right direction because it got access to the chat somehow (e.g. automated training that scraped his chat logs) or just a high level "Navier stokes can be solved through LLM". It sounds like the former may have happened although right now we just have an accusation and a weak denial.
That's also why they refused to answer that question: the answer is obviously "obviously"!
They explicitly say that they use your "content" to improve their models. Considering they practically have infinite compute at their disposal, why is it surprising that they would look for juicy data in there to make them look good ? When they ingested basically the entirety of human knowledge without regard to the rights of others, when they burn books by the thousands, when their relentless barrage of bots have rendered the Web borderline unusable, why would they stop at that line ?
Is it really hard to believe the people who would consume all the world’s data regardless of copyright and norms and permissions would not respect the data privacy of a user?
1. Less than a hundred people in the world are working at this problem, 2. A significant fraction of those happen to work at competing hyperscalers, 3. Those hyperscalers repeatedly show themselves not to take user privacy seriously
> Those hyperscalers repeatedly show themselves not to take user privacy seriously
where? Any examples?
yes.
Rules and laws are for the poor.
"Mr A. Nino weathers a storm of protest" - https://www.independent.co.uk/news/mr-a-nino-weathers-a-stor...
the paper makes a very serious allegation of dishonesty and possible academic misconduct.
the governance and integrity of openai is of importance to the welfare of society. this is not a matter of drama.
The strongest complaint is that they trained on a huge corpus of pirated copyrighted works.
It’s a large step above “scraping” and well into the “everyone acknowledges this is illegal” territory.
I think it's notable that nobody was calling out those researchers for their lack of integrity, because the systems they were building did not seem like a threat to anyone.
OpenAI etc get accused of a lack of integrity on this precisely because the systems they are building work, and are profitable.
My personal opinion here is that integrity is more about what you build with the data. I think saying "scraping means you lack integrity" is a simplification.
Then you have OpenAI etc.. who build these multi-billion (trillion??) dollar machines and sell them back to people, using everyone's proprietary data, and (among other things) tell everyone it's going to take their jobs. That combination of things doesn't scream integrity to me.
Still, it's undeniable that these machines could be beneficial for humanity (cancer research and such). So, I'm sure many people would say the good out-ways the bad. I don't know. Seems that would set a risky precedent for future companies, but maybe not.
OpenAI et al also stole everything from everyone. But then they raised billions of dollars from that data and sell back their LLM to people (again, among other things). They are also very much NOT open in any way, aside from sharing their benchmarks of new models.
legally speaking, the default privacy notice gives them an irrevocable license to your content. they may read and use the prompts for research. so it is very possible they simply stole the navier-stokes solution.
that is the same principle as any other prompt but this would be a concrete example.
there would be some difference between simply giving the model some prompts to read, which they are entitled to do on the default policy, and putting it into aggregate training data.
But the drama here is a little important. Stealing the millennium prize for N-S is sort of a big deal, especially to those who had been working on it for the last few years.
Particularly if the first proof being "solved" thanks to piles of money and compute for self-serving marketing discourages the mathematician who might have otherwise devoted years of focus to reach the superior proof we will now never see.
Tao agrees.
As a concrete example, such a proof could be less than a page with very specific initial and boundary conditions and inserting them into the equations to get something that goes to infinity when time goes to some finite value.
This would resolve the Millenium problem but not make humanity any smarter.
So if math is all that matters to you, you should care about this.
Regardless of mathematicians stating the methods outstrip the proof’s importance, still amazing we got an explicit social counterexample as well so quickly.
> 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.
The fact that this academic sniping can now be done at scale does change the formula though and shouldn't be ignored. The pressure to move math work into secrecy because at the slightest signal OpenAI and Anthropic will start burning tokens for headlines, is bad for math and its bad for everyone.
> In fact, it is now the identification of a promising problem which is the scarce and precious resource. We have now seen that even the rumor of someone working on a problem can trigger a massive amount of AI-powered effort to flatten it before the original research project has time to reach its full potential. The incentives may now be pointing in the direction of no longer sharing any promising research directions with the broader community, which would reverse centuries of traditions of open science and do serious long-term damage to the future of the field.
[1] https://mathstodon.xyz/@tao/117237322160500501
This is by no means new. Perhaps it is even more extreme now. Literally my first 1:1 with my PhD adviser back then, he told me that the most important thing about a researcher is the quality of the problems he picks.
In the real world the quality of these esoteric problems is typically gauged by the difficulty of solving them.
I'm pretty sure it would be considered plagiary amongst colleagues and it is a terrible precedent if we just let OpenAI steal any good idea they can get their hands on if they think it is profitable. You'd effectively sign away any and all rights to anything built with AI if OpenAI chooses to reengineer it before you.
I have terrible news about how literally every leading AI model was trained
Ai companies got where they are by stealing all of the intellectual property from human history. It seems entirely likely that their goal is to purloin everything produced going forward as well.
This is well-known to anyone in the industry.
Unless OpenAI finished a whole new training run on the latest data in the last few days, the possible allegation seems to be the latter.
They have been collaborating on this solution for a year, and Astra was trained in February this year so it’s entirely possible the direction of their research was in the training corpus.
They sure as hell don't need it just to produce English.
That article is only saying when you opt out there may be a loophole in the terms to allow OpenAI to train on the intermittent reasoning data anyways. If you don't opt out there is no ambiguity, all of the data can clearly be trained on.
So you have to opt out, it's just argued it's not clear from the terms that will also opt out of training on reasoning data or not.
The way for people or companies or universities to control their data and information is to keep it on their own computers.
"Do we use user feedback and de-identified data to improve ChatGPT and Codex in a holistic way? Yes. And so does every LLM company." - Mark Chen, Chief Research Officer, OpenAI.
https://x.com/markchen90/status/2097400166554993041
That “opt-out” thing is a dark pattern. It’s not a reliable and definitive way of protecting your data. Sometimes they flip on automatically when you accept a seemingly unrelated dialog box. Maybe you click it by mistake. You can’t take back what you’ve already shared. Also I don’t think it covers all the cases that they use your data. It’s really an opt-in button for voluntarily giving away your data for training.
This is covering for Tristan saying something like, "Actually, I was using my friend's account for half of this work".
Wildly disagree. "Training data" should not imply 'we can look at exactly what you are doing and then do it quicker and get the flowers for it', even if the terms allow for it.
I took their words as “can neither confirm nor deny”, in the that they are _presenting_ it as ambiguous, but I suspect it’s… less ambiguous to OpenAI.
OpenAI should release the agent log, including CoT.
This may sound like a charitable interpretation of OpenAI's remark, but consider that the lie would be (I think) impossible to falsify from the outside. They could easily just say "no sir we didn't peek" unless:
1. The conspiracy to peek at codex sessions involved enough people that the risk of one snitching is non-negligible
2. Lawyers advised it would be a bad idea to make such a remark, whether true or false
No; if they said "we can see that Tristan opted out of model improvement, therefore we are confident his work and ideas did not improve our model," that would be an excellent and reassuring precedent.
Yes, fair game, but innacurate to sell it in the media as an advancement of AI as some sort of artificial intelligence, and telling people to use the smart AI, when in actuality the mechanism by which the discovery was found was hybrid human/machine, and telling people to use this tool will result in the discoveries being sniped by the vendor.
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.
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 did not want to, declined to respond, and published first.
Also keep in mind we've only heard one side of the story, so any interpretation of events so far is incomplete. There should be a lot more information from OAI's side coming out later today.
"If you dont want me to be nice, then I dont have to"
-nice mobster
Either put up some evidence-backed arguments, or shut up.
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 the outrageous condition that he exclude his co-author from the credit?
OpenAI claims to already have a full proof (which they produced in the past 5 days after the rumors leaked). Hence the dispute.
What I find interesting is the timeline of when he found counterexample for NS is very unclear. Did Tristan find a counterexample weeks ago or was it very recently? Was it after OpenAI solved it? The wording is intentionally vague.
Either way, there was a massive rush to publish these results.
> 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.
This is such a sad mess, and it really didn't have to be this way.
https://x.com/dheeraj_nagaraj/status/2097266146445774924?s=6...
Whether the Codex sessions could have indeed made their way into Astra training data is something I can only speculate on though.
> I believe Luis Mart´ınez-Zoroa deserves a Fields Medal.
No looksies, wink.
No trainsies, wink.
[0] https://openai.com/policies/how-your-data-is-used-to-improve...
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.
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.
Also brain raping* is not illegal in most jurisdictions.
But they're both deeply disturbing.
_________
* https://youtu.be/JlwwVuSUUfc?si=uWl4-LCHAeI7qtb3
> 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.
I bet a lot of lawyers are salivating at this question too.
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?
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 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 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.
> 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.
> 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.
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.
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 would be shocked if they weren't tuning those models with the most relevant math texts and user material
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.
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
Model editing to remove PII that slipped through, all sorts of things of that sort.
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.
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
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.
> 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.”
And simply knowing a problem can be solved is half the battle.
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.
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 (...)
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.
EDIT: my understanding is that "solving the Navier-Stokes existence and smoothness problem" consists of proving at least 1 of 4 precise statements ("options (a) through (d) of Fefferman's statement of the problem"), and Luis and Diego's work were developments towards a proof of statements (c) and (d), which have been recently further expanded by Alpoge and Buckmaster
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)
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...
Have you done any mathematical research? If not, then no, knowing that a problem is solvable is not “half the battle”.
Homework problems are all designed to be solvable, yet they can vary greatly in difficulty. Research mathematics is even more extreme, because, unlike with homework, you don’t know that it is solvable with the extant mathematics, and you might need to invent new maths.
If not for the rumors that A/ had already solved NS, OAI would likely never have pursued solving the problem with such fervour. 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."
Let's see what statement OpenAI will come up with for their side of the story
EDIT: precised my thought on user data vs session data
"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."
"When a further trained version of our internal model became available over the course of the effort"
"While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models "
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?
That is not collaboration, and is not an academic norm.
That OpenAI heard about the result and decided to throw a lot of money at it knowing it was within reach ?
That the model took an approach that was published years ago ?
So either the rumors or Tristan's contact would explain it fine.
The whole point of Tristan's first email was to address the issue of the rumors so in fact this explanation confounds several things (in terms of the mutual knowledge of the conversants and their intentions).
Based on this lack of understanding it is pointless to continue this thread
Yes, I read fully the statement, what about you? Do you know what is a threat? What is this in your super-humble opinion if not a threat:
> 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.”
And you are saying that I don't have reading comprehension...
> 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.”
[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...
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.
2:47 am (eastern time for me) https://x.com/polynoamial/status/2097215233119211902
2:59 am https://x.com/_sholtodouglas/status/2097218331690578000/
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…
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.
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.
If they're going to try to beat researchers to discoveries like this it disincentives researchers to talk about their progress publicly, and basically breaks the ecosystem of scientific cooperation / discovery. It's also immoral.
So to say that it is unlikely is extremely suspicious. No, they did not literally pull user data. But user data is automatically added to their training set by default, so their latest in-house model would be trained on it if it is from several months ago. It isn't intentional on their part, and they probably realised they could not refute that they trained on Tristan's logs unintentionally, hence why they acted the way they did.
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 make sure your work will never been seen by anyone else because it will always be the ones making the “true advancements”.
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
Are those the same agents that a week ago escaped their sandboxes? How can OAI (the humans) vouch for agents they don’t - seemingly - have fully under control?
It's entirely possible OIA scrapers have picked up their work somehow, and then it was anonymized using some outsourcing effort.
> It's entirely possible OIA scrapers have picked up their work somehow, and then it was anonymized using some outsourcing effort.
Those two statements seem at odds with each other... Your stance is that they have enough insight into their agents behavior (leaving aside the agent sandbox escapes) that they can be certain none of the work was accessed but then conveniently don't have the ability to retroactively search the corpus of training data that they are feeding to this new model?
That seems convenient as fuck for OAI.
These are the kinds of people in charge of the reins, folks.
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 love to see more informed insight/discussion on this.
https://mastodon.social/@tristanbuckmaster/11723647135247030...
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)
"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.
As and aside here: With the HF incident and now this (suspected) one too, it seems that OpenAI may not have lost control of their bots, but it seems quite clear that they simply would not care even if they did.
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.
Skimming the PDFs it seems much more dramatic than that? It sounds like at least one of them is concerned OpenAI "solved" the problem by having their internal model use the chats of the independent researchers and want to claim the credit instead? I don't know. The tone is pretty accusational though:
> 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.
> 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. [0]
[0]: https://cims.nyu.edu/~tristanb/statement.pdf
If what he wrote is accurate, it does suggest that OAI is effectively extremely hostile to cutting edge researchers (eg, if we hear rumors about your partial success on a problem that has huge PR benefits, then we'll assemble a strike team of researchers with unlimited compute to claim the win for ourselves, possibly by training on your data). It's also not a good look for them to request author removals based on company affiliations.
I think what you have in mind is more appropriate for more normal corporate projects and the like. But academic collaborations are not usually so political/'profit' driven, if that makes sense.
Presumably because this was something Levent did in his spare time and because it was not obvious that this work would eventually lead to a breakthrough.
> Why did Tristan use OpenAI's models when it should have been known was a potential outcome?
I'm sure in the past he had less cynical feelings about OpenAI and their penchant for academic fraud.
> I understand they wanted a normal math collaboration but presumably what Levent brought was his resources (as far as I can see Navier-Stokes is not his speciality)
I think you're not giving the guy enough credit in saying that his contribution came down to having an API key for Anthropic models.
> Normally these things are hashed out formally beforehand to avoid the sort of thing now happening.
How would that have helped? That agreement (which may well still exist) would not have involved OpenAI.
While common sense reminds us here that if you send your data to an external entity’s computer, you are no longer in control of said data. The lines have blurred here clearly over the last decade, but that should have made the theory yet more clear to everyone involved: your data will be vacuumed up unless you keep it sealed. Use your own computer if you want to be in control.
I'd prefer things be opt in, and especially not start opt out, then try to trick you opt in with a popup defaulting to opt-in, like Anthropic did on consumer plans, but if they submitted anything on an opted-in plan it's not reasonable to be mad it trained on it.
Even still, I also believe for significant reasons that OpenAI would ignore the opt-out in selective cases and could be in the wrong here.
And the threats and terms they offered seem wrong either way, pending more context.
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.
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. Sholto Douglas from Anthropic is also on record saying the suggestion that OpenAI used Tristan's codex transcripts somehow is extremely unlikely to be true[1], which I agree with, though it doesn't rule out them learning of Tristan's work some other way. I am sure OpenAI will have a statement out tomorrow clarifying their position.
[1] https://x.com/_sholtodouglas/status/2097218240397410733
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?
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.
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.
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 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?”
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?
Where his tone is obviously corporate speak. "We heard rumours so we though we might give it a try, too!" as if (1) it wasn't FOMO that drove that decision and (2) perhaps that urgency would be a source of clouded judgment.
Not sure who I believe now, but it does seem like Buckmaster is just upset that NS is solved and not be his side.
That is what you would do if you wanted to beat someone to the punch.
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.
Then again, maybe this is my internal cope, hoping that they're not secretly training on private chats.
> 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?
"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
https://mastodon.social/@tristanbuckmaster/11723341370570119...
And here are Terence Tao’s comments on the results: https://mathstodon.xyz/@tao/117233527638291447
"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
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.
This is significant.
i always found it fascinating how existence and uniqueness of solutions for the basic types of PDEs (Laplace, wave, heat...) follows from boundary conditions of just the right type intuition tells us, i.e. either value or derivative for Laplace (corresponding to fixing voltage or charge on the conductors), both value and derivative for wave (corresponding to initial position and velocity of the parts of the string, as we'd expect from classical mechanics), and also something about the solutions for the heat equation being unstable for negative times (which totally makes sense when you think of "diffusion" -- can't unmix it).
Key quote : "Solving the problem by purely AI-powered methods [would be a] net negative for the progress of mathematics."
We are basically mass manufacturing math. Just like you have just 100 designers for a product selling millions of units, you will now need 100 mathematicians to make millions of advancement. Yes you have factory workers, but if we are being realistic they have negative leverage in the world and the analogue of that is not something most of today's mathematicians would want to do. They would want to be in the 100.
Like Tao says, each advancement is now significantly less useful since it yields fewer usable objects. However, we will get many many advancements. Is the tower made with many worse bricks better or worse than the tower made with a few amazing bricks? Depends on the tower. And time will tell.
For some fields of math and some of it's usecases, economies of scale will be positive ROI overall. In others it won't. But we will know which is which only after it's been fully scaled up, which will take 10-15y in my estimate.
Some feel that in the majority of usecases it is negative ROI, some feel the other way, but that opinion is for practicing mathematicians like Tao to hold. Also, some opinions on either side are held in the context of a particular field or practice, and should not be interpreted generally.
- 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.
Separately, Terence Tao noted there is a low probability OpenAI actually solved the general regularity problem.
It’s true that Tristan has no concrete evidence OAI accessed their session. It’s impossible for him to have that without OAI’s say so.
Why are you framing things so pro OAI?
- He insisted that OpenAI initiated research based solely on rumors and never accessed their Codex sessions.
- He mistakenly believed Tristan and Levent were solving the same problem in Anthropic.
- proposed two option. (1) Tristan becoming the lead author to revise OpenAI’s work, or (2) OpenAI providing internal model to support and bridge their research.
- Sebastian insisted there was no intention to alter authorship. He was simply uncomfortable sharing OpenAI’s work,model with an Anthropic researcher. Additionally, He believed Levent’s credit seemed limited as their work focused on Euler.
- complained that negotiations with Tristan and Levent were difficult
[1] https://xcancel.com/SebastienBubeck/status/20973794116915163...
Its not about accessing the session, its whether the session ended up in the training run for the next version of the model
That's normal practice for these models and it seems like that was the case since OAI added a disclaimer
Obviously they can't causally prove it helped though since these models are incomprehensible
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.
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
Terry also talks about it https://mathstodon.xyz/@tao/117234157753860650
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.
"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.
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.
https://www.quantamagazine.org/computer-helps-prove-long-sou...
- 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.”
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?
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.
$15m in tokens; but what about labor?
What about compressible fluids?
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
\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.
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.
While I support their argument - push for stronger data and privacy protections from OpenAI and similar - it is naive to believe we can have privacy while sending our data to third parties. It's clearly better to be safe than to be sorry here. Well, clearly better in terms of privacy. In terms of the maths gold rush, who can say what's better, that probably favours those taking more risk.
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
https://xcancel.com/ElliotGlazer/status/2096298696438906934#
That seems like a valid reason to contact OpenAI.
> 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 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.
Who still wants to use AI to solve cancer and other major problems?
(some drama from good ol' William)
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.
It seems like OpenAI heard of the rumor and then scooped them because their internal model is better/they have more compute. OpenAI has NO obligation to mention Tristan nor Levent, because they DID NOT steal their data.
Imagine if OpenAI opened up a high frequency trading arm and suddenly stole all the prompts and research that other HfT firms are doing through OpenAI tools and start making bank based on that . Wouldn’t that be straight up insane?