Comment by Kapura
1 day ago
> I have not actually read the paper. In fact, there are three closely related papers [TC26a], [TC26b], [TC26c], and I have not read any of them. I tried, but soon gave up. The math(s) dump does not appear to be strong on exposition. Instead, I asked Fable to explain [TC26a] to me. Fable seems to understand the argument and thinks that it is correct.
this does not feel like a healthy place to be
The very next lines say:
> At first I did not understand Fable’s explanations. But I asked questions, and then asked more questions, and drilled down towards the core of the problem. Gradually I started to understand parts of the argument. I have seen enough now to arrive at the tentative belief that the main result is possibly correct […] In the sections below I will try to explain what I have learned so far, and why I lean towards thinking it might be correct.
So the whole paper is about things the author does understand. (BTW the author David Harvey was behind the big advances in integer multiplication, in 2014 and 2019/2021.)
This is just textbook Gell-Mann amnesia. If the author doesn't read the paper how can he possibly know that? Even in this case when it's someone who has related knowledge, how can you know that you're being told something that's actually in the paper? LLMs confidently state nonsense all the time because there's no difference between how they produce a correct or incorrect answer. LLMs seem incredibly good at convincing people that they're now experts in everything. "Drilling down" is just giving more and more opportunities for that to happen.
- The author is already the world's best expert on this topic.
- You're right that it's possible Fable came up with a different o(n log n) algorithm that is not in OpenAI's paper, but (apart from being unlikely) it doesn't matter, as he talks of the belief that “the main result is possibly correct”, not necessarily that OpenAI's paper is: whether the argument comes from the paper or from Fable, the question of interest is whether multiplication is really possible in o(n log n) time. The point is that there's a genuine idea here, somewhere between OpenAI's paper and Fable's explanation of it and the author's understanding of it.
- This is in fact how mathematicians have evaluated papers for decades: is there any way this result could be true? Is there an actual idea here that could help? Do I understand enough of the argument to be convinced it's likely true, modulo some details that need to be filled in? The last step (actually filling in the details) happens when teaching it to someone else or writing it again in one's own words.
Gell-Mann amnesia doesn't really apply to the world expert on a topic discussing a paper that builds on his own work. It might apply to a world expert on a topic discussing a topic on which he is not an expert.
If I didn’t know David Harvey personally (though a decade removed) I would agree. I’m reconsidering my stance on AI lately and while I still find much of this disconcerting, I begrudgingly accept that his approach of "feed the paper into a different model and make it explain the technical bits until I fully understand it" is perhaps an increasingly valid approach to digesting mathematics. One needs to be incredibly principled in this practice. But that’s exactly how I’d describe David.
This clearly hit a nerve in very interesting ways!
Suppose we have a very convincing math PhD impersonator, agreeing and making some prepared smart-sounding comments saying they understood the proof and it seems to make sense - would it?
Well, most would say - NO! Obviously not, knowing this person was an impersonator. Others now knowing could be convinced of the validity.
Very simple thought exercise that allows to conclude, someone "saying they understood" (or agree) bears absolutely no weight to something's validity.
The idea that OpenAI's proofs are fake is contradicted by the response of mathematicians. Were they fake, mathematicians would not consider this release important enough to comment upon[1,2], as vixra always provided a steady supply of such writings. Nor would they be able to write expositions and simplifications, of which several are published each day on arxiv (just search OpenAI on arxiv to find them).
But perhaps mathematicians are no longer able to be the best judge if a proof is valid. Yet nobody but them is able to even understand what is claimed to have been resolved.
It is really strange to see mathematicians of the highest rank taking these results seriously[3], holding lectures explaining them, yet see it dismissed here as a mass psychosis. [1]https://smf.emath.fr/actualites-smf/20-10-26-communique-sur-... [2]https://www.mathsoc.jp/publicity/news20261016.html [3]https://terrytao.wordpress.com/2026/10/19/a-sketch-of-the-pr...
> But perhaps mathematicians are no longer able to be the best judge if a proof is valid
It seems they are still the best judge of whether a proof is useful or insightful though. AI is going to reward hack and try to eke out a win whatever it takes - which is fine if you regard these problems as trophies to be won, but not so fine if you take them in the spirit intended.
The famous problems that OpenAI (primarily) are trying to trophy-shoot with AI are famous because some human mathematician such as Erdos had the taste to judge them as fruitful grounds for new math to be developed in the pursuit.
Until we're at the point where AI is capable of things like creativity, curiosity and continual learning, AI will remain reliant on humans to fundamentally advance math and give it new tools to apply. For humans to do this it is helpful if they are inspired by the challenges set by their peers, which is why AI solving these human challenges is so counter productive - the snake is eating it's own tail.
1 reply →
> Were they fake, mathematicians would not consider this release important enough to comment upon[1,2], as vixra always provided a steady supply of such writings.
Few articles on vixra are made by companies with ~$1T valuations.
1 reply →
The truth is still tying it's shoes, in that metaphor.
Analysing all the proofs and finding contradictions or not takes time.
> But perhaps mathematicians are no longer able to be the best judge if a proof is valid.
Not the way they're written, and not at the rate they're dumped on the math community, at least.
[dead]
> Very simple thought exercise that allows to conclude, someone "saying they understood" (or agree) bears absolutely no weight to something's validity.
Well, it is always was so? It doesn't matter if the person claiming things is impersonator or not, either you dive into the topic yourself and decide for yourself, or you wait for a consensus among specialists to settle.
I mean, in 2026, after the rise of the Internet, where lies are widespread, AND after the rise of an AI, that removed such anchors of truth as a video proof for example, I believe everyone smart spend some time meditation on an operational definition of truth and now has something that doesn't depend on knowledge that the source is not an impersonator.
Yeah from an objective truth seeking perspective, and i agree with you, it's hard and often times impractical.
About 2y ago, before AI was all the rage - lookup: controversy, scientific fraud networks, "paper mills"
Now imagine that PhD “impersonator” has, in the past few months, made more demonstrable progress in mathematics than anyone in history.
I think that would change the conclusion of the argument a little.
Or maybe the PhD impersonator discovered a soundness bug in a component of Lean and was able to exploit it.
With no humans able to understand the proofs the impersonator has produced, there's no way of knowing.
7 replies →
More realistically, that impersonator has created lots of output, and the correct ones are selectively publicized (and few particularly interesting bad answers).
Yes from a social proof perspective.. not really from an objective truth seeking perspective, the best liar tell truths 99% of the time.
Do you really believe Open AI made more progress in math than any of the human predecessors? Come on.
Galois had built modern algebra before dying at 20. What agents did so far is flash in a pan.
5 replies →
The recent progress does not imply that the "impersonator" has a deep understanding of mathematics. By formalizing a problem in Lean, constructing the proof is essentially a complex search. Because Lean is the arbitrator of correctness, the "impersonator" can make as many mistakes as they want until they find a solution that passes the checker. That is very different from checking a paper for correctness.
9 replies →
The same is happening with AI-written codebase. You don't read AI written code, you ask an AI to explain it to you. Not that it's either a healthy place to be...
> You don't read AI written code, you ask an AI to explain it to you.
Rather: you reject AI-written code if it is not of really good quality.
In my experience dealing with people who use LLMs, approximately none of them do this. The feeling of productivity LLMs create is maximized if you just blindly accept everything they produce, and apparently as a field we've broadly given up on what little quality we had before to chase that feeling (and to boost the valuation of a couple companies).
> this does not feel like a healthy place to be
I believe you just captured the zeitgeist, friend.
Why not? If mathematicians in the field cannot understand the papers, but tools at our disposal can, why not use them? What is "unhealthy" about that? Just seems like it's the reality of the world now.
If you can't understand what the machine has done, math is not being done. Math is not pushing symbols around to find sequences that follow all the rules. Math is the process of understanding how those sequences of symbols map to meanings. If you can't follow the proof but have validated it's correct, all you know is that it's a proof of something. Is it a proof of what's claimed in the adjacent plaintext? Who knows!
Math is about clear communication between people first and foremost. That's why Shinichi Mochizuki's claimed proof of the ABC conjecture isn't accepted - he's unable to communicate it to enough mathematicians to get a consensus on its validity. LLMs generating Lean proofs are basically following that same schema. Sure, something has been proven. But what? Where's the clear communication explaining it to the satisfaction of other mathematicians?
> Math is about clear communication between people first and foremost.
I can't imagine I'm the first one to point out that this is perhaps the greatest unsolved problem in mathematics.
> Where's the clear communication explaining it to the satisfaction of other mathematicians?
Coming in about a year for $20/month, sure seems like.
3 replies →
Human understanding is important, but I don’t think it’s the end all be all. There’s no reason to expect humans to be the best mathematical reasoners in the universe. And if math is important, beyond being a human art form, I think we should explore that math beyond our understanding.
2 replies →
"Mathematics is commonly explained and recorded in symbolic and concrete forms that are easy to communicate, rather than in conceptual forms that are easy to understand once communicated. Translation in the direction conceptual -> concrete and symbolic is much easier than translation in the reverse direction, and symbolic forms often replaces the conceptual forms of understanding."
https://mathoverflow.net/questions/43690/whats-a-mathematici...
Does math exist without humans?
I mean, I could not understand 95% of all the math out there due to skill issue, and it’s still valid. If machines can understand the math and humans can’t, we can’t validate it, but that doesn’t make the math itself invalid.
20 replies →
Why does this not feel like a healthy place to be?
The author of those sentences is also the author of the main result that this generated improvement builds on, and prior to this improvement his work was thought to be unimprovable. He is, in all of the important ways, literally the single most qualified human on the planet to asses this result. And his assessment is 1) this improvement is real, 2) this improvement is exponentially bigger than his work and the work he improved on, 3) it is effectively impossible for him (and by implication, humanity) to understand the result as presented, 4) it is seemingly trivial for “frontier circa 2026”-class models to understand the result as presented, and 5) to the extent that he is able to understand it, it is only with significant frontier model assistance.
So the place we are at, at least for this highly intellectual field, is “intelligence is the property of ~2 corporations, you must pay them a fee and agree to their terms and conditions to access it”.
At risk of stating the obvious, this is not a healthy place to be because intelligence is now structurally subordinate to the profit motives of particular corporations and at the mercy of all the attendant economic/sociopolitical incentives that apply to them. If the landscape is such that it is in these corporations interests to withhold this intelligence from some and make it available to others, that is simply what will happen.
One trivial example, of which there are an arbitrary number: An elderly leader is convinced to issue a decree that all number-math-computation intelligence is to be made available only to the intelligence agencies of his nation, and failure to comply means no IPO and no further investment. From this point on there would be a permanent nation-state-shaped shadow over practically all modern cryptography, with no recourse possible.
If the Church had a Large Astronomy Model in the time of Copernicus, the earth would have stood still - regardless of what it may privately prove about heliocentrism, it would never declare e pur si muove.
Leaving aside the (much more important) issue of monopolizing intelligence, and assuming the claimed result is indeed correct, the question is whether a) the correctness proof is indeed too complex to be understood by humans, or b) the creator (the LLM) is not able to write it down in a form digestible by humans.
Observing the LLM output on problems that have clear solutions invented by humans, I tend to believe that too many people are jumping to conclusions too soon and interpreting b) as a).
If the LLMs are truly so great, we should see (in a short time, i.e. in order of months) a transformation of the math proofs generated by LLMs from the current state of indigestible mess to something that indeed looks like written by human for humans.
Only then would one be able to claim that we got into situation a).
5 replies →
> An elderly leader is convinced to issue a decree that all number-math-computation intelligence is to be made available only to the intelligence agencies of his nation, and failure to comply means no IPO and no further investment.
Who would staff those intelligence agencies if humans are no longer able to understand the work product?
1 reply →
Yup, and it always has been, we don’t even understand how our bodies or minds work, yet we use them everyday
The amount of unimaginable complexity contained in a seemingly very tiny amount of matter to us, is incredible, and it’s all around us in nature
We’ve never had the capacity to fully comprehend neither ourselves nor the environment we live in. But we do our best, and we do survive and sometimes even thrive
well math has always been about "understanding things" -- hence proofs and etc is the main tool in the field
(and why math differentiates between conjectures (maybe true) and theorems)
so that "oh some 3rd party person/machine understood it for me" can feel... a bit icky
3 replies →
IDK why this is relevant to the topic at hand.
Everyone in advanced math research understands the topics they are reading about deeply.
This person is just literally being victimized by the AI incompetence apocalypse.
Information overload + trained laziness.
The mistakes that lurked in radium, in arsenic, in leaded gas, in fallout downwind of nuclear tests ... all had the best tools that the previous generation knew.
None of them were okay. Those were small small tools with few (sometimes only one) uses.
Generations earlier, famines, diseases and lead-leeching plumbing made rulers insane and entire cities of people die without any help from the "best tools".
Scifi is always written about the best tools. The issues arrive in the tools and in the hands that wield them both.
There's nothing that makes this paper the be and and end all opinion pieces of the paper. The community will review his review, and people who wish to understand more will read more and more of the paper. I see nothing wrong with this process of progressive understanding.
What a funny pov?? Mathematics and other pure sciences are about human understanding. You're saying if humans cannot understand the thing that was for human understanding, that's ok? That's like saying all the world's food is now inedible, but that's ok because a blender can consume it still.
One day AI might use all these complex math to build a ftl rocket and you will still be what's the point if we can't understand it?
3 replies →
The "top layer" of maths being this is one thing. Now imagine another layer of this. Then another. Then another. Then another.
At one point intuition helps to build out usage of the math itself. See stuff like "oh I can do rotations through matrix multiplication"-style ideas.
It is helpful for us to be able to reach for those tools quickly. The more tools that require advanced prep to use, the less we'll be able to use the results in easier contexts. Everything will be hard
Of course nothing really prevents people from looking for the simpler tools and proofs. Lots of math stuff starts off complicated and gets simpler.
Just important that this is still a goal in mind
There are a few increasingly powerful companies that own the frontier models that can produces these results. In the future, it may only be these models that understand the results. Should we be beholden to a few companies for basic understanding of foundational principals?
Because the tools don't understand the papers either. We know for a fact that:
1. These tools are wrong a significant % of the time
2. They have an upper complexity bound where they start to fall apart
Relying on AI to check AI papers is a terrible idea
So, just like humans then?
But where they do differ is they do have an infinite stamina, and can work on problems in loops, self-correcting their own mistakes if a formal way of checking the correctness of the solution exists. Which does happen to be the case with math.
8 replies →
Yes you're replying to astroturfers or AI psychosis victims.
Tools don't "understand" anything. They put together a probable string of tokens.
As you did ^
We don’t actually understand what “understanding” means. We’re just uttering plausible strings of words around the topic.
1 reply →
> Why not? If mathematicians in the field cannot understand the papers, but tools at our disposal can (...)
How can you tell?
Especially for tools built to tell us we’re right and happily pretend to solve problems they didn’t. I think the OpenAI “revert” of their math problems speaks volumes of how volatile this can be.
I'm not a mathematician but from what I've heard from them that's generally not how mathematics works. You can "use" whatever you want, of course, but a mathematician's goal and main activity is finding and understanding proofs. The ones I knew also liked compiling error lists for math textbooks and sending them to the authors. (It opened my eyes because I admit to have been one of those ignorants who implicitly assumed "it's formal stuff in a book, sure it's going to be right".)
The tool may malfunction. Indeed there are theorems that say the tool must malfunction at least sometimes.
If you don't understand the subject and relies on the understanding of the tool, how will you have a frigging chance of catching an error?
Anyway this reminds me that scene in Matrix 2, where the mayor (of sorts, don't remember his exact title) of Zion tells Neo about the machines that keeps everyone alive, but that there is nobody left that understands how they work or could repair them if they broke.
>Indeed there are theorems that say the tool must malfunction at least sometimes.
Gödel's incompleteness theorem only states that there are true statements that can never be proved to be true. That limits what tools can do, but it's not a "malfunction". The exact same limitation applies to humans. A genuine malfunction would be if the tool "proved" a false statement to be true. That actually has happened before because of bugs in proof checkers, but software bugs can be found and fixed, and proof checkers are simple enough that it's possible to write many different implementations. It's extraordinarily unlikely that the same malfunction occurs in all of them.
Sounds like a disassembler.
because the tools can't? Lol. That's not how it works. Am I in the twilight zone?
All this anthropomorphism of literal statistical text generators.
Yes they exhibit generated sentences of text that describe coherent responses about a topic when there's an inordinate amount of research material in the training data from which they are distilling sentences of sound and rational thought. That's the ONLY way it can possibly work. There are so many scenarios where the training data isn't setup like that and low and behold the generated responses are just literal garbage.
Honestly this is starting to feel like a "God in the gaps" argument.
3 replies →
Fable is the "gadget" of language ELI5.
Honestly it is not that much different from the divine inspiration if you have ever experienced it.
I do not know how you view the world or walk through it but language in general is simply subject to interpretation to me, the arrangement of words for the sentence is inspiration in and of itself - though the majority of time it amounts to minimum jumps.
Fable was able to help a human understand the paper though, so the situation is quite a bit better than if we just had a lean proof that was apparently correct but for which all attempts at explanation were "hallucinations". There doesn't seem to be an apparently obstacle to a human fully understanding the proof in several months of focused study.
This is typical of working with agents, not only maths. "I implemented this with agent A, and don't understand the logic, but agent B think its correct". But then you ask a simple clarifying questions to B on a random thing in the code and it'll tell you that this part is actually wrong. At least in maths they have Lean...
This is a conundrum I hadn’t seen anyone predict. We will reach a time when machines push math and physics farther than a human mind ever could. But we won’t be able to know if it is correct for that same reason!
Math suddenly becomes biology
I've seen this claim repeated a bunch of times, it spreads as a meme, but the analogy doesn't work that well. This is not a special feature of biology! Chemical engineering involves lots of trial and error, physics is all about neglecting complicated aspects of a situation and hoping the result is still useful, even fundamental physics has a bunch of cancelations that work with no very good justification (see troubles around renormalization) and then fail when you try to do quantum gravity. Even math has things we know are true but have no good conception of why -- see John Conway on the Monster Group ("there's never been any kind of explanation for why it's there"): https://m.youtube.com/watch?v=xOCe5HUObD4
The whole world has always been like this. It's just showing up with force here.
Why, some morons have been summoning the Singularity for quite a while. Ask them what they're gonna do in that Singularity and get an answer along the lines of "pursuing hobbies".
There's a huge lot of suicidal people who ideate their afterlife as "pursuing hobbies".
Who thought four years ago or even two years ago that we would be in a place where these models are coming up with solutions that our best minds have a hard time comprehending and can’t do it without their help. What a time to be alive.