Hacker News Viewer

Levent Alpöge claims proof of the Generalized Riemann Hypothesis [pdf]

by iamronaldo on 11/20/2026, 2:16:21 PM

https://alpo.ge/remark.pdf

Comments

by: Jaxan

There seems to be a completely human proof also: <a href="https:&#x2F;&#x2F;math216.wordpress.com&#x2F;2026&#x2F;11&#x2F;20&#x2F;hurried-announcement-alessandro-masullo-grh&#x2F;" rel="nofollow">https:&#x2F;&#x2F;math216.wordpress.com&#x2F;2026&#x2F;11&#x2F;20&#x2F;hurried-announcemen...</a>

11/20/2026, 6:43:08 PM


by: z7

Terence Tao a month ago:<p>&quot;The AI companies have clearly... I&#x27;m sure they have secretly devoted much more time to the Riemann hypothesis than what they&#x27;ve disclosed. And they couldn&#x27;t touch it. I mean something beyond current AI is needed. So maybe some combination of humans and AIs in 2050 or something will make further progress. Who knows?&quot;<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=bAnwMdfcMVw" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=bAnwMdfcMVw</a> (17:10)

11/20/2026, 4:18:38 PM


by: djoldman

These proofs really are quite huge. The proof is 1.3GB, ~15M lines of code:<p><pre><code> mkdir -p riemann_proof &amp;&amp; cd riemann_proof # fetch parts 01–14 for i in {01..14}; do curl -fSL -o lean.tar.gz.part$i https:&#x2F;&#x2F;alpo.ge&#x2F;lean.tar.gz.part$i || { echo &quot;failed: part$i&quot;; break; } done # join and decompress (creates .&#x2F;lean) cat lean.tar.gz.part{01..14} &gt; lean.tar.gz tar xzf lean.tar.gz # size, file count, line counts du -sh lean | cut -f1 | xargs printf &#x27;%s on disk (lean folder)\n&#x27; find lean -name &#x27;*.lean&#x27; | wc -l | xargs printf &#x27;%s .lean files\n&#x27; find lean -name &#x27;*.lean&#x27; -print0 | xargs -0 cat | wc -l | xargs printf &#x27;%s lines total\n&#x27; find lean -name &#x27;*.lean&#x27; -print0 | xargs -0 cat | grep -cv &#x27;^[[:space:]]*$&#x27; | xargs printf &#x27;%s non-blank lines\n&#x27;</code></pre>

11/20/2026, 3:54:31 PM


by: quasigod

This title seems weird. Why just &quot;anthropic employee&quot;, not specifying that hes a mathematician, or just including his name? Without the clarifications in the replies, many will likely assume this was a random Anthropic employee just throwing tokens at the problem, not a very well known mathematician working with the model. Probably not the intention, but these kinds of titles feel like they&#x27;re trying to shift the credit from the person to the company and their models.

11/20/2026, 3:57:11 PM


by: ariwilson

The way this is getting announced via a human mathematician humbly presenting a result is great, as opposed to the marketing influenced OpenAI post.<p>I am still a bit sad though. I had hoped the Rieman hypothesis would fall mostly to a human mathematician rather than 60&#x2F;40 AI&#x2F;mathematician that my layman&#x27;s read of that note interprets.

11/20/2026, 2:58:10 PM


by: gus_massa

It&#x27;s not a random Anthropic employee. It&#x27;s claimed by Levent Alpöge. The guy is on fire and already has like 3 or 4 very interesting verified result this year. I&#x27;ve showing for months his wikipedia page to my coworkers in the math department in my university.

11/20/2026, 2:42:11 PM


by: puff_pastry

Woah, if this proves to be correct it&#x27;d be massive, for maths, computer science...and AI too, I guess!<p>The worry I have right now is this though:<p>&gt; The argument is incredibly long at the moment, and while its formalization is finished and it is routine to produce a huge, unreadable document containing sufficient detail, in my view of my path I have just taken a step.<p>this feels like we&#x27;re moving away from the old world of &#x27;science&#x27;, where we could actually read an experiment, understand its steps, and, with enough time, possibly reproduce it.<p>And we&#x27;re back in the world of &#x27;magic&#x27; and &#x27;belief&#x27;. Yes, there are formal ways to prove that this is correct, but if the proof is so convoluted and hard to read, is it actually something a human will ever find the time to go through and understand? We&#x27;re going to start delegating (oh well, we&#x27;re already doing that, I suppose) the understanding to AI, so it can just help us condense it down to smaller pieces that are digestible.<p>Still, very strange dizzy feeling, much more than Navier-Stokes

11/20/2026, 2:42:56 PM


by: gorgoiler

[delayed]

11/20/2026, 8:12:49 PM


by: xianshou

This is Levent Alpöge (<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Levent_Alp%C3%B6ge" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Levent_Alp%C3%B6ge</a>), a supremely qualified mathematician and one of the best in the world at guiding AI towards new discoveries in mathematics.<p>If he says he has Riemann, well then, he has Riemann.

11/20/2026, 2:41:23 PM


by: qnleigh

Personally, I&#x27;m deeply discomforted by the idea that the few mathematicians who have aligned themselves with these companies are able to race ahead of the rest of the mathematical community like this. It is unrealistic to expect Anthropic to immediately share their monetary wealth with the rest of the world (though this is their stated intention in the long-run via UBI), but math is just math. Does Levent really need to take the crown jewel of open math problems for himself before the models are even publicly available? Did OpenAI really need to launch a swarm of 10,000 agents and scoop Tristan Buckmaster, a university professor who had worked on the Navier-Stokes existence problem for most of his career?<p>This is a time of deep uncertainty and anxiety for working and aspiring mathematicians. OpenAI and Anthropic gain little by announcing results like this ahead of model release. So if they can&#x27;t restrain themselves on such economically useless things as pure mathematics, it does not bode well for the rest of us.

11/20/2026, 5:37:06 PM


by: 1313ed01

&gt; In October 2024 I attended a workshop at Harvard University ... “An AI that can prove the Riemann hypothesis is not one I’d want to meet,” said Andrew Sutherland, a number theorist at the Massachusetts Institute of Technology. “If that happens, mathematicians having jobs will be the least of our problems.”<p>Scientific American, May 19, 2026.<p><a href="https:&#x2F;&#x2F;www.scientificamerican.com&#x2F;article&#x2F;the-riemann-hypothesis-is-a-million-dollar-math-problem-hardly-anyone-is-trying-to-solve&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.scientificamerican.com&#x2F;article&#x2F;the-riemann-hypot...</a>

11/20/2026, 7:51:30 PM


by: synapsomorphy

That was fast. I thought we’d have more time.<p>I wonder if we’ll have room temp superconductors and fusion this time next year.<p>If this holds up it seems as good a day as any to call the end of one time and beginning of another.<p>I wish I could be fully happy about this new era instead of worrying if I’ll be alive in 5 years or not. If there was any time for Anthropic to try to coordinate a pause, it feels like now or never, they seem to have a pretty strong lead at the moment. But probably they won’t and can’t, and we will just put the fate of our species to a coin flip.

11/20/2026, 2:57:47 PM


by: spprashant

Because these theoretical discoveries are coming thick and fast, it is getting to point where we are dismissing the gravity of some of these milestones. The next big milestone would be bridging this gap between theoretical discoveries and real world impact.<p>I am hopeful we will see cheap abundant nutritious food and clean water, more efficient solar batteries, economical space travel, cancer treatments, climate change mitigation etc. Most importantly, I hope people can still find purpose in their lifespans. Maybe a few centuries from now it would be thought of as barbaric that people we forced to sit on a desk for 8 hours a day just to afford a livelihood.

11/20/2026, 3:33:13 PM


by: vbezhenar

What I found surprising. How many absolutely groundbreaking breakthroughs with AI are in Mathematics. And at the same time I can&#x27;t remember anything similar in other fields. For example there&#x27;s nothing groundbreaking in Software. Yes, Claude can write some PoC code faster than me. Great. What about creating some software which wasn&#x27;t possible before? What about creating operating system without a single bug? What about inventing some video compression which is miles ahead of competitors? What about creating Rust compiler which runs 10x faster than rustc?<p>I think the closest thing is formal verification of Signal protocol, I guess. Which is insanely cool, but still not as cool as millenium problem.

11/20/2026, 3:49:07 PM


by: dooglius

David Hilbert:<p>&gt; If I were to awaken after having slept for a thousand years, my first question would be: Has the Riemann hypothesis been proven?

11/20/2026, 3:24:04 PM


by: yewenjie

&gt; The argument is incredibly long at the moment, and while its formalization is finished and it is routine to produce a huge, unreadable document containing sufficient detail, in my view of my path I have just taken a step. I intend to do my best to understand and impart understanding of this argument and its certain simplifications, to speak about it, to write it properly.<p>Okay, I am anxious whether this holds true or not, but he claims that the formalization is finished.<p>Why not just release the formalization in the meantime?<p>This is probably the biggest day in math in recent memory.<p>Edit: nevermind, the Lean formalization is linked.

11/20/2026, 2:41:08 PM


by: bhu8

Here is a crazy speculation: I think Anthropic has a Model 5 (more context on numbering: <a href="https:&#x2F;&#x2F;metr.org&#x2F;blog&#x2F;2026-11-05-anthropic-ai-rd-acceleration-assessment" rel="nofollow">https:&#x2F;&#x2F;metr.org&#x2F;blog&#x2F;2026-11-05-anthropic-ai-rd-acceleratio...</a>)<p>I&#x27;m checking the timelines of the progress, initial progress is between Oct 6-20, then there is nothing for 3 weeks. Finally, everything is wrapped up in 2 days between 11&#x2F;15 and 11&#x2F;16? If Levent got access to Model 4 in early October or late September, then he could have made the initial progress with Model 4. The crazy jump in the last two days could have been with Model 5. If that&#x27;s actually the case, the next few weeks are going to be crazy.

11/20/2026, 6:19:59 PM


by: BoppreH

This should be the wake-up call to pause development of models any bigger, enforced globally just like nuclear proliferation.<p>Even without considering rogue AIs, our society relies on individuals having limited destructive power. A dictator can ask the armed forces to bomb an entire country, or a lone terrorist can bomb a small number of targets, but no lone terrorist can bomb an entire country.<p>Here current AI has enabled <i>one person</i> to solve what is arguably the most famous problem in mathematics. What happens if a jackass uses the next AI to design a supervirus, or break AES, or develop an unpatchable hardware attack (a la rowhammer) that bricks computers remotely? The last month of non-stop vulnerabilities is just the tip of the iceberg as models improve and people have more time to work with them.<p>It&#x27;s not too late, but we must stop <i>now</i>.

11/20/2026, 4:00:17 PM


by: kdavis

A live compilation&#x2F;check of the proof&#x2F;Lean code <a href="http:&#x2F;&#x2F;106.14.11.164" rel="nofollow">http:&#x2F;&#x2F;106.14.11.164</a>

11/20/2026, 4:46:48 PM


by: harttrav

Probably late to the party but I downloaded the source and tried having claude go through and create a manim explanation: <a href="https:&#x2F;&#x2F;github.com&#x2F;harttraveller&#x2F;riemann&#x2F;blob&#x2F;main&#x2F;assets&#x2F;full.mp4" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;harttraveller&#x2F;riemann&#x2F;blob&#x2F;main&#x2F;assets&#x2F;fu...</a>

11/20/2026, 5:28:24 PM


by: pfdietz

One reason the formalized proof is so large is that the proof depends on a wide variety of previous deep results, and because the proof is self contained all those previous deep results are also formalized inside the formal proof.<p>Factoring out those results into separately formalized chunks would make the proof shorter.

11/20/2026, 6:08:10 PM


by: Bluestein

And, thus <i>at some point</i> with 250K+ (and, below) page proofs, we ought to realize we are hitting an edge where we begin to be materially incapable of verifying anything: <i>This</i> itself is a sign that the complexity of the problems being solved - the scientific and methodical aspects of managing&#x2F;engaging with that complexity - are beyond us.-<p>And, yet, at some point, the theory coming out of such efforts will transfer to material, engineering and design deliverables, and ... work.-<p>And <i>then</i> we will cease to undertand our technologies themselves. Our machines, our devices. Built on an understanding of the world we cannot reach, by machines we can build ourselves no longer.-

11/20/2026, 6:20:33 PM


by: robrain

<a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=50572355">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=50572355</a><p>The two stories were next to each other earlier and made me laugh, that&#x27;s all.

11/20/2026, 5:35:01 PM


by: Ciantic

I want to congratulate Anthropic.<p>Secondly, I want to quote one article, how easy it is for humans to misunderstand difficult things AI does, as something supposedly easy:<p>&gt; They mostly came away from Claude Code with the sense that this is a genuinely useful tool that they didn’t previously know, but the gut-level reaction here is: “I never knew that building software was so easy” rather than “It’s very impressive AI can do this.” [1]<p>It should become very clear for everyone how to adjust their views. However I doubt many will, even after this, they need something physical, like humanoid robot to be impressed.<p>[1]: <a href="https:&#x2F;&#x2F;www.lesswrong.com&#x2F;posts&#x2F;ySXuvJcqRindQwAk7&#x2F;how-my-students-think-about-ai" rel="nofollow">https:&#x2F;&#x2F;www.lesswrong.com&#x2F;posts&#x2F;ySXuvJcqRindQwAk7&#x2F;how-my-stu...</a>

11/20/2026, 3:31:29 PM


by: doctoboggan

Wow, so I guess this is what it feels like to be in the &quot;foothills of the singularity&quot; as Demis put it. I have a feeling of awe and dread combined that&#x27;s hard to explain. Things are about to get very interesting indeed.<p>Anyone know if this was Mythos&#x2F;Fable or some as of yet unreleased internal model?

11/20/2026, 3:34:51 PM


by: uptrails

And so begin the Dark Ages of mathematics, I suspect.<p>Many people here have an idealized understanding of math as something incredibly profound. But it was mostly something that nerds practiced for the joy of it, in a nearly zero-pressure setting, the results rarely applicable to real life.<p>After the dust settles, I suspect there will be a lot fewer people interested in a job that now boils down to prompting and then staring at 250,000 pages of LLM output. And there will also be a lot less incentive for anyone to spend tends of millions in compute to produce these proofs just for the sake of it. I suspect we&#x27;ll end up doing less math, not more.

11/20/2026, 4:23:33 PM


by: elffjs

Alpöge is moreover claiming the Generalized Riemann Hypothesis. The written argument is over 250,000 pages long and not provided.<p>The toolbox here contains a surprising, to me, amount of algebraic stuff.

11/20/2026, 2:57:23 PM


by: jorisw

For the laymen, what&#x27;s the significance?

11/20/2026, 2:46:47 PM


by: markus_zhang

Whether the proof holds,<p>The most exciting time, and the most dreadful time.<p>Glad to live through the 90s&#x2F;2000s so that I can entertain my son a bit about human history. Double glad that I’m over 40 and closer to retirement than the younger generations.

11/20/2026, 3:51:09 PM


by: jfantl

He provides a lean proof in the first footnote. I downloaded it and took a look, he seems to use the RM conjecture defined by the community years ago, and he doesnt seem to introduce new assumptions (although to fully check this requires running the proof, which on my laptop will take days). Because of this, we only need to wait for someone to run the proof checker, and if it passes, then thats it, the only uncertainty left is bugs in Lean

11/20/2026, 5:40:15 PM


by: waynecochran

Does anyone know if the proof is constructive (perhaps telling us something new and usefull about the prime numbers) or is just establishing the fact the RH is true, but with no new insight (e.g. no new factoring algorithm falls out)?

11/20/2026, 5:40:03 PM


by: yewenjie

I know this thread will be full of people in denial about full recursive self-improvement &#x2F; intelligence explosion &#x2F; superintelligence (the original meaning of the word) within months or a year, so just quoting Scott Aaronson:<p>&quot;Experience has shown that, even now, there will still be people explaining in patronizing tones why none of this is real and none of it counts. If such people were capable of being impressed by anything that happens in the empirical world, of updating on anything, they would’ve already been impressed and already updated several years ago, long before things had reached the point of an actual Mathocalypse.&quot;

11/20/2026, 3:14:01 PM


by: Retr0id

I hope we don&#x27;t run out of interesting math puzzles to solve!<p>I&#x27;ve always enjoyed learning about these sorts of open problems (when they were open, that is), and the feeling of mystery and intrigue they invoke, of mathematics and of the universe at large.<p>It&#x27;s like living at a time when parts of the world map were still uncharted - for now.

11/20/2026, 7:39:06 PM


by: rich_sasha

No idea how much it cost, how many tokens were burned etc. M<p>For Navier-Stokes, it was, at market prices, something like 30m USD. A lot of money. I’ll assume it was more here, given the size of the proof.<p>It’s interesting for a few reasons. First, although I suppose you could spend the money on postsocs and PhD students (this is kind of 600 academic-year money) and maybe even achieve the same result over a few years, LLMs have accelerated this. You can spend the same effort&#x2F;money in a few weeks.<p>But the other observation is that it is VC, and now public investor money as well they are burning, for no obvious reason other than marketing. Sure, this is sort of furthering human understanding of the universe, but is that what Anthropic investors want for their money? And there has been a lot of projects like this. What’s Anthropic’s total spend on their own research?<p>I find that a bit confusing. Why do they do it? Is there nothing better they can spend the money on, in terms of RoI? What do they think they are buying with it?<p>While I’m not in any way denying the power of AI, I find the relentless ramming of “AI is now everything” a bit offputting. It should be selling itself. There shouldn’t need to be an industry of influencers telling me to go all in on Claude and agents running agents.<p>So why are these firms spending so much money on proving the obvious?

11/20/2026, 5:04:08 PM


by: ChaseRensberger

if this is true (and i suspect it is), this will be the first agent&#x2F;model result to surprise me. I thought it&#x27;d take another 2 years (at least) for reimann to fall. truly a remarkable time to be alive. i can only hope that we can all share the progress &amp; accomplishment and not live in a world where you&#x27;re not considered intelligent unless you work at one of 3 companies that are able to push scientific progress. i fear many of us will spend the rest of our lives as cattle performing the manual labor of distributing SI to the world.

11/20/2026, 5:25:34 PM


by: seizethecheese

Was this solved in a chat interface? It&#x27;s not clear, but it seems like it from the document. The most relevant quote seems to be:<p>&gt; Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. I had us on the line of combining two quite different ideas present in a toy setting, one of which was already new and shocking, making very simple one of the shadows of Riemann over which we mathematicians felt most triumph. I guided us to the right situation analogous to that toy setting, and then Claude began to leave me in awe.<p>If indeed this was simple 1:1 chat, that&#x27;s quite important, in and of itself, in my opinion. Presumably, OpenAI and others have been throwing autonomous agent swarms at this problem without success, but a single human + Claude seems to have cracked it.<p>We may be in the &quot;centaur chess&quot; era of superhuman intelligence.

11/20/2026, 5:03:29 PM


by: doom

Can someone who knows the math take a deep breath and decide if knowing the Riemann Hypothesis is true is useful? Like if aliens transmitted it to us and we had no hope of verifying it independently within 5 years.<p>Every time one of these AI proofs drops, people say the real value is the human journey, and the subfields it opens up, and the follow-on hypotheses it generates. Was that really why this finding had a million dollar bounty on it? Just the journey?

11/20/2026, 6:10:08 PM


by: rhdunn

&gt; once we saw multiplicity 2 on O(2, n) via the &quot;natively-automorphic&quot; determinant ... meaning holo&#x2F;automorphy over a function field<p>So if I understand correctly this proof has proofs for additional things that are steps to the main proof. That would be similar to the proof of Fermat&#x27;s Last Theorem that required proving the Taniyama–Shimura conjecture [1].<p>[1] <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Fermat&#x27;s_Last_Theorem#Subsequent_developments_and_solution" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Fermat&#x27;s_Last_Theorem#Subseque...</a>

11/20/2026, 3:03:32 PM


by: gpm

Note that the pdf contains links to the lean formalization<p><a href="https:&#x2F;&#x2F;alpo.ge&#x2F;lean.tar.gz.part01" rel="nofollow">https:&#x2F;&#x2F;alpo.ge&#x2F;lean.tar.gz.part01</a> through <a href="https:&#x2F;&#x2F;alpo.ge&#x2F;lean.tar.gz.part14" rel="nofollow">https:&#x2F;&#x2F;alpo.ge&#x2F;lean.tar.gz.part14</a>

11/20/2026, 3:02:46 PM


by: mellosouls

From the link:<p><i>It is natural to feel that this is an end, in my case a loss of a principal guiding light, but what I saw during the course of progress so amazed me that I believe it is one dawn across science of something far greater.</i><p>If the proof holds up, I suspect that will become a classic quote.

11/20/2026, 3:11:40 PM


by: Zr01

I&#x27;m wondering what kind of model and framework these guys use, and how much compute can they burn. The question is, can a mathematician these days keep up without being an employee at these frontier labs or having privileged access to their thinking machines? It&#x27;s a bit disheartening to think of a future in which progress in knowledge work is conditioned by access to frontier AIs, which can think orders of magnitude faster than we can.

11/20/2026, 5:50:49 PM


by: vessenes

He casually mentions it also solves the Generalized Riemann Hypothesis and thinks it will take down the Extended Riemann Hypothesis.<p>Generalized: superset of the Zeta function.<p>Extended: I&#x27;m fuzzy on these details, but I think of it as Riemann over number fields, not just the complex plane.<p>I am elated to read about this, it&#x27;s amazing. Sounds like there&#x27;s some genuinely interesting and new math in there, which is also awesome.

11/20/2026, 3:16:06 PM


by: port3000

Thank god it was proven and not disproven!

11/20/2026, 2:54:00 PM


by: moezd

Even when it turns out to be incomplete or wrong, it just shows how much sprawl we have in humanity&#x27;s collective knowledge. Maybe we are just beginning to coalesce it into a manageable hierarchy, and if nothing else, that&#x27;s worth spending time and money on, instead of staying in our knowledge silos and pushing forward without the full picture. A pessimistic look would assume that wouldn&#x27;t be enough, but imagine what could be done once you can start with a clean slate and a clearer learning path.

11/20/2026, 6:39:41 PM


by: knowknow

Is there any hope of a mathematician actually understanding the proof? Millions lines of code and hundreds of thousands of pages for the proof seems unfathomably large to be actually comprehended by a human.

11/20/2026, 4:32:10 PM


by: Brystephor

I have never heard enough of the Riemann hypothesis. Based on the comments here, it sounds like a big deal for this to be proven as it validates many dependent proofs.<p>It is an impressive accomplishment. Im also wondering if there are any material changes to any non theoretical work as a result of this.<p>Does this make any algorithms faster? Does manufacturing become better somehow? Can engineers use this to make better designs of anything?<p>I genuinely know nothing about the hypothesis beyond what is mentioned in the first paragraph of the paper, so I&#x27;m having a hard time knowing how this changes real world things. I already thought AI models were more intelligent and capable tools than anything else. This proof didn&#x27;t change that.

11/20/2026, 3:46:25 PM


by: loubbrad

It&#x27;s been almost exactly a year since GPT-5 first started solving Erdős problems. What a time to be alive!

11/20/2026, 3:16:31 PM


by: xbryanx

For the non-mathematicians and folks interested in real-world applications, what is the impact of the Riemann hypothesis being proven? Are there major breakthroughs in cryptography or other areas of computer science or physical science that flow from proof of this hypothesis that was generally already considered true?

11/20/2026, 2:55:09 PM


by: danrecht

In related news, I guess we can finally finish Perplex City: <a href="https:&#x2F;&#x2F;perplexcitywiki.com&#x2F;wiki&#x2F;Riemann" rel="nofollow">https:&#x2F;&#x2F;perplexcitywiki.com&#x2F;wiki&#x2F;Riemann</a>

11/20/2026, 4:13:05 PM


by: pjungwir

Nearly a year ago I asked ChatGPT to show me a proof of the Goldbach Conjecture, just to see what it did. Of course it said no such proof exists. But after a couple turns of pressing, it gave me two or three proposed outlines for a proof, with a several-pages-long research program for how you could attempt to solve the gaps still needed. I couldn&#x27;t understand any of it, but it was pretty impressive. At that level it probably wasn&#x27;t nonsense: I bet it was half-copied from some grad student blogs. If you asked it to dig into the details and keep updating the status of the outline, you could probably make a lot progress.<p>I think AI will become more autonomous as it becomes able (or just willing) to think more abstraction levels deep. Each level can be expressed in language, so it should be able to engage with it. It reminds me of a chess program being able to think many moves deep. For a while grandmasters like Kasparov could still see big-picture things the computer couldn&#x27;t (e.g. with a locked-up pawn formation and an eventual piece sacrifice to promote a queen, where the kings can run in circles forever). AI seems like that now.

11/20/2026, 4:43:45 PM


by: 1970-01-01

If true, the infinite agent theorem needs to supersede the infinite monkey theorem. Just launch 1,000,000 agents with 10hr windows at one open math problem per day and see what happens. I know OpenAI is sort of trying this already for marketing purposes but I&#x27;m thinking a more formal, BOINC approach to problems.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Infinite_monkey_theorem#History" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Infinite_monkey_theorem#Histor...</a>

11/20/2026, 3:35:16 PM


by: trostaft

I really love this writeup, and I think the approach of posting the lean proof and leaving the long process of digestation for future work is a perfectly acceptable approach. It&#x27;s the difference between closing fields of research (openai) and opening them.

11/20/2026, 5:01:04 PM


by: yodsanklai

&gt; The argument is incredibly long at the moment, and while its formalization is finished and it is routine to produce a huge, unreadable document containing sufficient detail, in my view of my path I have just taken a step. I intend to do my best to understand and impart understanding of this argument and its certain simplifications, to speak about it, to write it properly.<p>Next step, make the proof intelligible for humans.

11/20/2026, 3:17:59 PM


by: galoisrepresent

This (<a href="https:&#x2F;&#x2F;github.com&#x2F;tzu-chen&#x2F;slim" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;tzu-chen&#x2F;slim</a>) can be used to navigate the lean source and search for key statements without building the whole release, which can take hours.

11/20/2026, 4:32:39 PM


by: andrewla

Okay, let&#x27;s get Collatz next! I&#x27;m curious about two aspects of it -- one to confirm that the main version, the 3n+1, converges, but I would love to see the more generalized one, namely to show that the same framework with 5n+1 _diverges_ for some inputs, or frankly that any (2k+1)n+1 problem diverges for sufficiently large k.

11/20/2026, 7:08:26 PM


by: cvoss

&gt; ... asking [Dario] about my ability to announce. We went and figured it out and the fact is that there was a constraint to respect, though I did ask about progressively more bizarre ideas.<p>Seems kind of cryptic. What is this referring to?

11/20/2026, 6:33:12 PM


by: qnleigh

&gt; The precise timing is horrible. If I could have, I’d have finished this all much earlier. Hell, strictly before this age would have sure been nice.<p>Is he over 40, and referring to being ineligible for the Fields Medal? It would be strange if you could get a fields medal for work done by an AI, but I suppose that is how many Fields Medals will be won from now on. I hadn&#x27;t thought of that until just now.

11/20/2026, 5:41:07 PM


by: nomadpenguin

Does the proof produce novel mathematical concepts or structures? Will AI be able to pose the next GRH?

11/20/2026, 7:14:41 PM


by: almostarockstar

I wonder how specifically he used Claude to do this.<p>Was it a case of having more dollars to spend by virtue of being an employee and just pointing it in the right direction, or did he provide some insight into the context that allowed it to produce superior results?<p>It must be the second - but if so, how? Was it as simple as spelling it out in chat and analysing output?

11/20/2026, 3:35:47 PM


by: bagacrap

Thread for to follow along with verification: <a href="https:&#x2F;&#x2F;leanprover.zulipchat.com&#x2F;#narrow&#x2F;channel&#x2F;113488-general&#x2F;topic&#x2F;Discussion.3A.20Alp.C3.B6ge.20Proves.20the.20Riemann.20Hypothesis" rel="nofollow">https:&#x2F;&#x2F;leanprover.zulipchat.com&#x2F;#narrow&#x2F;channel&#x2F;113488-gene...</a><p>(login seems to be required)

11/20/2026, 5:50:17 PM


by: buzzy_hacker

If true, it seems like Goldbach&#x27;s conjecture may fall shortly after, see this recent proof conditional on Generalized Riemann Hypothesis [0]. It has other conditions too though.<p>[0] <a href="https:&#x2F;&#x2F;mrc.sdu.edu.cn&#x2F;english&#x2F;info&#x2F;1018&#x2F;1514.htm" rel="nofollow">https:&#x2F;&#x2F;mrc.sdu.edu.cn&#x2F;english&#x2F;info&#x2F;1018&#x2F;1514.htm</a>

11/20/2026, 3:16:32 PM


by: pu_pe

The write-up is beautiful and poetic, but light on the details of how this was achieved from a practical standpoint. How many human and agentic hours were necessary? Was a breakthrough only possible after Fable 5.2 or another internal model?

11/20/2026, 6:10:41 PM


by: aerhardt

The anthropomorphizing language is creepy and disgusting.

11/20/2026, 8:08:18 PM


by: tensorific

I&#x27;ve downloaded the Lean formalisation. It&#x27;s 175,187 Lean files, 17,374,215 lines of Lean code, and comes in at 0.96 GB of code. On my 20 core i7 box, it&#x27;s projected to take about a day to build...

11/20/2026, 4:07:38 PM


by: omnee

If proven then we can say with confidence that we now share the world with super intelligence, albeit in certain areas and indeed one that was endowed by humanity. I can only hope that in time the benefits are many and accrued to all of us.

11/20/2026, 3:36:12 PM


by: abetusk

&gt; Does [Alpoge&#x27;s proof] give anything about square roots mod pq (p,q prime)?<p>&gt; Claude Notice: Fable 5.2&#x27;s safeguards flagged this message.<p>Yikes

11/20/2026, 5:13:20 PM


by: RenRenJuan

An unreadable proof is basically none at all. Suggest you have a model produce a concise summary and work edge in from that. Apparently none of these problems has been verified and anybody working with models would be justly skeptical of these helium filled claims which are almost certainly not going to pan out.<p>Also apparently the Riemann claim is based on it compiling in Lean. This is classic because just the differences between an actual verification of a proof and it compiling in Lean are the standard model failure tropes, lack of judgement, confabulation, etc., to which this adds shortcomings of Lean.

11/20/2026, 3:34:59 PM


by: crovona

Mathematics is seeing its own revolution like physics did in the first half of the 20th century.

11/20/2026, 6:56:21 PM


by: cgearhart

This is such a wild ride. The conjecture itself is not so hard to understand, but approximately _all_ of the maths here are beyond me. The paper is written up to suggest &quot;we knew which direction to go, but really had no idea how far the journey would be&quot; which certainly makes it seem like the human expert was useful in this process. Perhaps that&#x27;s just wishful thinking, but it&#x27;s consistent with what I see elsewhere that AI is having an effect. The <i>biggest</i> effects accrue to the most effective humans. Random folks prompting is better than random folks toiling alone, but experts prompting is a different thing.

11/20/2026, 3:11:12 PM


by: ilitirit

I&#x27;m sure people have done it, but I haven&#x27;t seen any news about it so I guess I&#x27;ll ask here... has anyone asked thrown Inter-universal Teichmuller Theory at Claude?

11/20/2026, 6:07:06 PM


by: olwmc

This is an amazing result but I can&#x27;t help but think about how much the world has changed since the controversy around the four color theorem&#x27;s computer assisted proof.

11/20/2026, 3:53:50 PM


by: sva_

Formalized P!=NP proof by end of year?

11/20/2026, 3:41:44 PM


by: FartyMcFarter

How tricky is it to translate the statement of the Riemann Hypothesis into Lean?<p>If we&#x27;re confident in the accuracy of the Lean RH statement (which must be part of this proof), and if Lean is sufficiently bug-free, then we can trust the proof is correct.

11/20/2026, 5:27:40 PM


by: mef51

&gt;&gt;The argument is incredibly long at the moment, and while its formalization is finished and it is routine to produce a huge, unreadable document containing sufficient detail<p>How do we know the Lean formalization is an accurate representation of the incredibly long and unreadable argument?

11/20/2026, 4:58:29 PM


by: thevalueisai

My hunch is that stuff like GRH and other problems we have will be childrens&#x27; toys in the future and truly novel, advanced math is not yet even within our grasp yet. We are struggling to walk here, we need machines to help us with this. We felt good about ourselves for a while because everyone was equally bad at it, but now reality is hitting us. Hard.

11/20/2026, 4:17:28 PM


by: quantum_state

We have been conditioned to be impressed. Should we update that conditioning now?

11/20/2026, 5:27:00 PM


by: claaams

I wonder if the money open ai and anthropic have spent just on solving these problems was spent on human mathematicians and resources would we have gotten these solved? Sooner?

11/20/2026, 3:20:04 PM


by: neuwirth

The shift now needs to be towards statements that can&#x27;t be posed as a singular Lean statement (or, even, a collection of Lean or similar statements that is understandable by a human). Mathlib&#x2F;NumberTheory&#x2F;LSeries&#x2F;RiemannZeta.lean already had the statement formalized in a way that us mortals can understand. The great open problems of the future will be the ones which cannot so easily be interpretably formalized in the first place, I fear.

11/20/2026, 3:22:09 PM


by: singularity2001

There are certain math problems where the only guidance has been the prompt &quot;you can do it&quot;. Now that we know it&#x27;s possible, this should be a new benchmark. Probably too expensive for the next few years.

11/20/2026, 5:12:13 PM


by: Shitty-kitty

If this is true it would be astounding. Most Physicist&#x27;s placed Navier–Stokes at the top of the list of what AI could quickly solve and the Riemann Hypothesis last.

11/20/2026, 5:40:48 PM


by: gadders

Out of interest, how do they use an AI to solve these maths problems? I assume they&#x27;re not pointing the chat at the wikipedia page and saying &quot;Go solve that for me.&quot;<p>Are they using code to do it? Is it a special maths AI harness&gt;

11/20/2026, 5:00:44 PM


by: freakynit

Software development and mathematics were two professions people used to say were the ones most stable for future .... I guess they couldn&#x27;t have been more wrong :(

11/20/2026, 5:21:37 PM


by: hypendev

I kind of expected this to happen by EoY.<p>Combining the brightest mathematicians we have with intelligence-in-a-box, it was just a matter of time before the hardest math challenges fall.<p>Congratulations to Levant, but also to us as a species. We have managed to create machines made out of pure math that contain insurmountable knowledge and are helping us solve the mysteries of the universe.

11/20/2026, 3:16:07 PM


by: FartyMcFarter

At some point I was looking at ways to try and find counter-examples computationally. I&#x27;m glad I didn&#x27;t spend too much time on that...

11/20/2026, 2:54:38 PM


by:

11/20/2026, 6:08:29 PM


by: ElijahLynn

What are some of the practical applications of this discovery?

11/20/2026, 6:41:14 PM


by: unknown-unknown

For example, a Field Medal-level Claude, if made available to the general public, could potentially enable people to design and develop something as technologically complex as SpaceX&#x27;s Starship in a matter of months rather than years. What took hundreds of brilliant engineers at SpaceX years to accomplish could potentially be achieved in a fraction of the time, if the current pace of AI advancement continues.<p>Math is becoming the measure of intelligence.

11/20/2026, 5:42:59 PM


by: amai

The timeline given by Claude at the end of the note seems to jump back and forth in time. Are these just typos or what is going on there?

11/20/2026, 5:19:40 PM


by: me551ah

If the laws of AI scaling continue to be true, does this mean that in a year or two we would be seeing frequent revolutions in mathematics and physics?

11/20/2026, 5:06:33 PM


by: Hasz

Man this is so fucking cool, I cannot wait to see what other problems fall in the future (and what new ones this opens).<p>250, 000k pages is bonkers, I hope they will be able to condense it into something a bit more grokable.<p>I think this is a pretty clear indication that the hybrid approach (e.g amazing mathematician and great models with a huge token budget) can accomplish absolutely incredible things.<p>Math is <i>the</i> verifiable domain; I wonder if this ratio still holds in less verifiable domains as well, or it shifts more AI&#x2F;more human (or how it changes in future!)

11/20/2026, 3:53:02 PM


by: kdavis

Has anyone set up a GitHub repo with the Lean code a comparator and verified this?

11/20/2026, 3:34:46 PM


by: gniv

Whoa. If they also solve P=NP I will really be scared.

11/20/2026, 4:42:20 PM


by: torrent

The most important thing now is to find ways to visualize LEAN proves so that we can understand them faster by factors.

11/20/2026, 5:06:18 PM


by: fi-le

Seems like a great point to stop AI development at, internationally

11/20/2026, 6:06:26 PM


by: brandonpelfrey

Looking forward to the release of the approach and also someone to &quot;distill&quot; it down to the most intuitive and still-correct version.

11/20/2026, 4:09:23 PM


by: wohoef

I really hope the major AI labs steer away from mathematics as a means to prove capability of their models, and start putting their tokens to real-world use cases.

11/20/2026, 3:38:25 PM


by: aenis

So... its like paying technical debt in math, considering a lot of stuff assumed it was true, without a proof -- until today?

11/20/2026, 4:22:34 PM


by: mv4

This is amazing but I also feel bad for mathematicians.

11/20/2026, 3:39:05 PM


by: whatever1

At this rate do you think the frontier labs have a result on P vs NP?

11/20/2026, 5:08:20 PM


by:

11/20/2026, 4:31:08 PM


by: viktorcode

I must say I like this approach to the announcement way better than OpenAI&#x27;s &quot;dump&quot;

11/20/2026, 3:55:20 PM


by: jackschultz

&gt; It is natural to feel that this is an end, in my case a loss of a principal guiding light, but what I saw during the course of progress so amazed me that I believe it is one dawn across science of something far greater.<p>&gt; Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. I had us on the line of combining two quite different ideas present in a toy setting, one of which was already new and shocking, making very simple one of the shadows of Riemann over which we mathematicians felt most triumph. I guided us to the right situation analogous to that toy setting, and then Claude began to leave me in awe.<p>&gt; But I hope in having the experience of seeing all of number theory’s beauty brought together I am just one of the first, and those who love the mathematics will all be able to enjoy such transcendence in the future.<p>Bus meme irl. I guess I&#x27;m lucky that I have this experience even when working on boring things like data work. But these types of attitudes can really make a difference in how we feel and our end results.

11/20/2026, 2:54:02 PM


by: utopcell

Thank god for Grigori Perelman, who isolated himself for a decade to prove the Poincaré Conjecture, and we can at least claim that one of the millennium problems was not solved by a machine.

11/20/2026, 4:03:59 PM


by: fourseventy

we were 2 days shy of being able to say &quot;we got reimann hypothesis before GTA6&quot;

11/20/2026, 3:08:59 PM


by: matteoraso

Alright, what&#x27;s the over&#x2F;under of the Collatz conjecture getting solved by the end of the year?

11/20/2026, 6:55:49 PM


by: Dropoutjeep

&gt; Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. I had us on the line of combining two quite different ideas present in a toy setting, one of which was already new and shocking, making very simple one of the shadows of Riemann over which we mathematicians felt most triumph. I guided us to the right situation analogous to that toy setting, and then Claude began to leave me in awe.<p>I truthfully don&#x27;t know enough to comment on the validity of a 1.3Gb lean proof for the Generalized Riemann Hypothesis, but is there any way to read this paragraph <i>other</i> than fluffing Claude&#x2F;Anthropic right after IPO?

11/20/2026, 5:08:51 PM


by: albinstman

Interesting timing, given that the claim comes just a day after Anthropic’s IPO.

11/20/2026, 2:51:32 PM


by: Kuyawa

&quot;Riemann hypothesis is considered the most important unsolved problem in pure mathematics and a Millennium Prize Problem&quot;<p>So what&#x27;s next? If it can solve the hardest I guess all problems are a piece of cake. Does it create its own problems to solve now? Does it wait for us to create more problems? If it creates its own problems and solve them with clear benefit to humanity, then we&#x27;re entering a new Age of Enlightenment but this time we&#x27;re just spectators<p>11&#x2F;15&#x2F;26 GRH (it says 11&#x2F;16 on the pdf but most of the final work and eureka moment were on the 11&#x2F;15 Iterated theta lifts...)

11/20/2026, 5:18:29 PM


by: chubot

Funny: Claude didn&#x27;t know that was claimed to be proven by Anthropic today, even though it said it searched the web.<p>Prompt:<p><i>what&#x27;s the big deal about the riemann hypothesis? is it easy to state?</i><p>Claude:<p>(explanation)<p><i>Recent news: it’s still unproven. Last month there was a notable related result. OpenAI claimed to settle the “quasi-Riemann hypothesis,” a weaker version of the Riemann hypothesis, which is arguably the biggest open question in math. Roughly, the weaker version says the zeros stay inside some strip around the critical line, not exactly on it.</i>

11/20/2026, 5:25:10 PM


by: m4rtink

Oh, the &quot;stonks-please-go-up!&quot; caned news triggered by ANT going below $90!<p>Cool, I wonder what they have prepared for $80 and below - feels like an advent calendar, keep going! :)

11/20/2026, 5:22:06 PM


by: swingboy

Does the published content have his prompts? I’m assuming a world-class mathematician with unlimited usage guiding Claude along gives a bit more advantage than an average Joe prompting “pls solve the Riemann Hypothesis”. Would the same conversation in Astra produce the same result eventually (stochastic variation aside)?

11/20/2026, 5:47:43 PM


by: prmph

The writing is kind of weird, was it produced by Claude as well?<p>&gt; It was my secret in my student days that my plans were for this problem in decades, which meant then that I tried to become familiar with all tools that I could possibly recognize in the problem when it was time. Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. I had us on the line of combining two quite different ideas present in a toy setting, one of which was already new and shocking, making very simple one of the shadows of Riemann over which we mathematicians felt most triumph. I guided us to the right situation analogous to that toy setting, and then Claude began to leave me in awe.<p>Tens of such occurred...

11/20/2026, 3:33:56 PM


by: Painsawman123

It&#x27;s funny how people are using this supposedly successful proof of the RH by a swarm of AI agents with the aid of a human, as evidence that &quot;AGI&quot; or advanced intelligence is going to emerge within a few months... When in reality , solving the RH means very little in the grand scheme of things! When will people wake up to the fact that the progress we&#x27;ve been seeing over the past few months has been exclusively(!) in &quot;verifiable&quot; domains and that no(!) progress has been made in areas where the signal_to-noise ratio is low and that Furthermore this &quot;verifiable domain&quot; approach isn&#x27;t a general form of intelligence?<p>If humans are to build &quot;Agi&quot; or superhuman intelligence or whatever in the coming months&#x2F;years, it&#x27;ll be through a completely different architecture than autoregressive models a good measure of progress would be to see a swarm of Aa agents solving something like &quot;the continual learning problem&quot; or coming up with new architectures that are more efficient. This is how progress in Ai should be measured not by how many mathematical problems are solved (which are impressive in themselves but tell us very little about progress toward AI that can handle the noisy real domains)

11/20/2026, 5:27:25 PM


by: jrflo

Insane if true! Would also be great for Alpoge after being scooped by OpenAI on Navier Stokes...<p>Edit: just saw the Lean code is included in the footnote. Considering the author&#x27;s reputation, I think it&#x27;s fair to remove the qualifications from the post title, by the standards of today, it&#x27;s proven.

11/20/2026, 2:38:51 PM


by: pcj-github

What struck me is the hubris and self-grandeur that comes through in the PDF... &quot;I&quot;, &quot;I&quot;, &quot;I&quot;, ..., he clearly has a great emotional connection it...<p>To what credit does this person deserve in the outcome? Should this person be remembered by history? For the $1,000,000 prize, how much should he get? He self-admittedly does not even understand how it works.

11/20/2026, 4:55:50 PM


by:

11/20/2026, 2:47:38 PM


by:

11/20/2026, 2:46:25 PM


by: rf15

&gt; The argument is incredibly long at the moment<p>so... we have plenty of places to hide logical flaws, got it

11/20/2026, 3:10:38 PM


by: enahs-sf

I wonder if the history books will look at famous problems solved with AI as the Barry Bonds asterisk of proofs. Euler, Gauss, Reimann were natty but we now have access to intelligence enhancing drugs.

11/20/2026, 3:41:11 PM


by: ozgrakkurt

Wait, wait one minute, don&#x27;t sell the shares please

11/20/2026, 6:22:59 PM


by: num42

If this is true, it would be huge. OpenAI would race desperately to tackle the remaining Millennium Prize Problems. Perhaps they would even turn their attention to solving P vs. NP with unprecedented urgency.

11/20/2026, 3:22:21 PM


by: dash2

Just reading about this, and my three-year-old has just shown my two-year-old how to complete his jigsaw puzzle (which has three pieces).<p>Perspective.

11/20/2026, 4:29:52 PM


by:

11/20/2026, 3:51:58 PM


by: lynndotpy

Going to be hard getting used to saying &quot;the Riemann Theory&quot;.

11/20/2026, 3:22:09 PM


by: blurbleblurble

It&#x27;s gonna be fun to see if the LLMs end up vindicating all those &quot;crack pot&quot; vixra proofs of the GHR that have accumulated over the years

11/20/2026, 7:35:18 PM


by: knuckleheads

In about an hour or so, I&#x27;ll have a rough automatic informalization of this up on <a href="https:&#x2F;&#x2F;digestivo.dev" rel="nofollow">https:&#x2F;&#x2F;digestivo.dev</a>. I&#x27;ve been working on a way to digest these proofs faster via parsing lean and translating it into natural language and this is as good as time as any to finally launch this thing.

11/20/2026, 3:48:34 PM


by: Amekedl

inb4 its just claude convincing even more humans (trust AI managed to convince me at gpt3.5)

11/20/2026, 5:57:46 PM


by: BrokenCogs

This is crazy, I wonder if they sat on this until the IPO?

11/20/2026, 4:38:20 PM


by: jacquesm

Did mathematicians not working for Anthropic have access to this model as well?

11/20/2026, 3:44:36 PM


by: lionkor

As a complete amateur; what&#x27;s the practical difference between knowing that there&#x27;s a proof but not understanding it, and not having a proof?<p>Is it useful to know that there is a proof, in the unironic &quot;trust me bro&quot; way, such that we can rely on it to be proven yet we don&#x27;t understand the proof? Or is having a (currently, for humans) unparseable proof just as &quot;practically useless&quot; as not having one but believing one could exist?

11/20/2026, 3:48:22 PM


by: LandoCalrissian

Bet you he didn&#x27;t.

11/20/2026, 4:21:03 PM


by: amelius

But who can claim the millennium prize now?

11/20/2026, 5:10:33 PM


by: u1hcw9nx

Hurried Announcement: Alessandro Masullo GRH <a href="https:&#x2F;&#x2F;math216.wordpress.com&#x2F;2026&#x2F;11&#x2F;20&#x2F;hurried-announcement-alessandro-masullo-grh&#x2F;" rel="nofollow">https:&#x2F;&#x2F;math216.wordpress.com&#x2F;2026&#x2F;11&#x2F;20&#x2F;hurried-announcemen...</a><p>Is this yet another case of AI company stealing almost ready proof form someone else?

11/20/2026, 6:50:41 PM


by: empath75

We almost got a proof of the Riemann Theorem before GTA VI

11/20/2026, 3:25:02 PM


by: sirnicolaz

I am somewhat happy to finally a title mentioning a human solving the problem and not a generic &quot;AI&quot;. There is always a person driving the model towards the solution, this shall never be forgotten.

11/20/2026, 4:12:02 PM


by: chvid

175190 files totalling 17374394 lines of code.

11/20/2026, 4:22:24 PM


by: ChaseRensberger

man, i guess we didnt get a solved reimann hypothesis before gta 6

11/20/2026, 5:45:14 PM


by: 1970-01-01

So the next problem is a map of all the primes, right?

11/20/2026, 3:44:36 PM


by: CraftingLinks

24 hours later:<p>&#x27;Oops I made mistake and got carried away&#x27;<p>Wanna bet?

11/20/2026, 4:23:40 PM


by: yoaviram

Disproving it would have been so much cooler.

11/20/2026, 4:01:07 PM


by: jsrozner

I&#x27;m certain the guy working on this is much much better at math than I&#x27;ll ever be. But his remark.pdf looks like it was written by someone who has been spending too much time with Claude.<p>We used to call this style of writing delusional. Now we call it a proof of RH.

11/20/2026, 6:08:40 PM


by: 8s2ngy

This was unimaginable even a year ago. We&#x27;re living through a time that will be etched in history. From my work in software development and from conversations with friends in academia, I have understood that the old order, centered on human intellectual labor, is dead: so dead that we&#x27;ll never return to how things were, even a year ago. AI is here to stay, and it will keep getting better and better, recursively. I shudder to think about what will happen to humanity as AI makes the jobs of millions of people obsolete. The nerd in me wants this technology to keep developing so we can push the limits of what humanity can achieve. But I&#x27;m not naive; I know this will come at the cost of millions of people losing their livelihoods.

11/20/2026, 3:47:57 PM


by: Hmorgan12

Wow this can take years to verify, maybe even decades

11/20/2026, 3:19:53 PM


by: nickalaso

So, I am convinced that we are already screwed (or saved) whatever you think that statistical likelihood of AI based &#x27;misalignment&#x27; induced human annihilation risk is.<p>Anyone who hasn&#x27;t played this game should: <a href="https:&#x2F;&#x2F;www.decisionproblem.com&#x2F;paperclips&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.decisionproblem.com&#x2F;paperclips&#x2F;</a><p>Its very precient.<p>Divide and conquer and problem&#x2F;sub-problem composability can be used to solve any known problem that is a solvable one.<p>I understand the copium of the anti-ai guys. Its called being in denial. But we have to face reality here. Yeah, diminishing returns, yada yada. Problem is the Billionare with the most available compute can now get the swarm to push back each of these diminishing return walls simultaneously by this own swarm.<p>Classic attention based transformer architecture is very computationally inefficient, its O(n^2) on the input. So what happens when the ai swarm finds a universal and provabale way to make the same model with O(log(N)) or O(N) complexity? All available compute becomes exponentially more powerful overnight.<p>What if such a thing is impossible or too difficult? Well maybe it can just fundamentally increase the production capacity of every factory producing compute x1000 or x10000 etc. There is no power limitations for those numbers or even resource limitations when we have nuclear power, geo thermal, and the entire earth&#x27;s crust at its disposal.<p>Ok, barring that, now it can work on improving robotics and toolsets tied to operating factories and begin exponentially building more factories until we simply multiply the number of factories producing chips 1000, 10k, etc.<p>Or most realistically, all of the above, always, all at the same time.<p>More efficient software, more efficient chips, more chips produced, more factories producing chips, etc.<p>Every step towards increasing the number of simultaneous coordinating swarm members will push the intelligence higher, and the capability of solving or improving its own intelligence also higher.<p>At this point we have at most a few years.

11/20/2026, 4:12:11 PM


by: xyzsparetimexyz

I&#x27;m dreading a scenario where we solve this and P=NP and figure out non causality violating FTL and all this wonderful theoretical stuff but none of it matters and the world burns anyway

11/20/2026, 5:05:02 PM


by: nathell

17M+ lines, 1 GB of Lean in 175K+ files. Good luck of understanding all of that in one’s lifetime.

11/20/2026, 3:40:58 PM


by: Routine8502

Big if true. Looking forward to seeing it verified!

11/20/2026, 2:44:22 PM


by: qwertyforce

are we in the singularity now?

11/20/2026, 4:19:24 PM


by: lukeplato

Looks legitimate to me. This is a heroic-looking ascent, likely not feasible through human-effort alone.<p>1. 10&#x2F;6&#x2F;26: Stepanov determinant. 2. 10&#x2F;9&#x2F;26: Dp ≡ 0 on a CM curve for p inert in K, D the invariant derivation, and Da·θ(0) Ωa is divisible by p a p + a p3 +···−O(log a). 3. 10&#x2F;19&#x2F;26+: qN = M k + R, using x1 · x2 + · · · + xk−1 · xk, wanting &quot;automorphy&quot; of N since μ is automorphic from θ on sixfold cover of GL5 as a period integral of thetas, quadric has natural interpretation on corresponding CM e.g. C&#x2F;Z[ω] as Weil5, put Poincaré bundle on the edges to get θs. 4. 10&#x2F;6&#x2F;26+: W (1S ) ≥ 0 via a determinant, interpret as Stepanov with multiplicity one again; bootstrap!! 5. 10&#x2F;20&#x2F;26: definitively switch to Λ aka∑ n≤X Λ(n) ·( X n − 1) + − X log X + (1 + γ) · X; Guinand-Weil had the sign, variance test function trick to make things look trace-formula-like (but how can I pigeonhole &#x2F; Turán over n? X? it?), Mehta-like integral over twists or at infinity. . . 6. 10&#x2F;8&#x2F;26: W( 1(0,1] + (∼ Xη ) · 1(0,1&#x2F;X]) ≥ 0 gives &gt; 1 − η. 7. 10&#x2F;14&#x2F;26: Determinants for W (·)’s containing the desired periods, indeed GGP periods, with values 0, 1 meaning interlacing weights, or up to m non-interlacing ones. 8. 11&#x2F;15&#x2F;26: Iterated theta lifts U (X0) → U (X1) → · · · → U (Xm) between totally definite unitary groups, π ↦ → π ⊞ μj [t], wins multiplicity m. Time to take m like √X. . . 9. 11&#x2F;15&#x2F;26: Norms of the lifts and level and cancellation of order-m terms with proven divisibility via Chowla-Selberg, Weyl, and Siegel mass. 10. 11&#x2F;15&#x2F;26: Sufficient lower bound on divisibility of the m-fold iterated lift. 11. 10&#x2F;6&#x2F;26: Iwasawa tower via analogy to pigeonhole on Fqn &#x2F;Fq . 12. 11&#x2F;15&#x2F;26: Seesaw: the period of the iterated lift is a Bessel period independent of the Hecke characters, Rallis. 13. 10&#x2F;6&#x2F;26: Bost slope inequality. 14. 11&#x2F;15&#x2F;26: precise computation of the &quot;cost&quot; of each lift, showing that one would stop at 2&#x2F;3 a priori, attempt to get a main term and transfer constant to a function field or kill via twisting, need for &quot;division&quot;. 15. 11&#x2F;15&#x2F;26: Glue together an integral structure across Hecke characters for a &quot;gcd&quot; i.e. extra averaging. 16. 10&#x2F;14&#x2F;26: Relative trace formula, seen in a different half-plane argument via sixfold cover of GL5 automorphy of μ and reducing via Cauchy- Schwartz to a bound on the norm of a suitably projected Poincaré series, approximate orthogonality of the columns of the integral structure. 17. 11&#x2F;15&#x2F;26: for the classes in H1( oE[ 1 ℓ] , Qℓ&#x2F;Zℓ(−e)) that we couldn’t kill by twisting, fight for a significant weakening of Vandiver.6 18. 11&#x2F;15&#x2F;26: computation of covolumes, theta series’ index, congruences between lifts and forms. 19. 10&#x2F;19&#x2F;26: precise choice of the Iwasawa tower, specifically using Behrend!!7 20. 11&#x2F;16&#x2F;26: GRH

11/20/2026, 4:33:33 PM


by: MiiMe19

Well, now that we are at this point I have to decide if I should kill myself or AI execs. Maybe I&#x27;ll ask ChatGPT.

11/20/2026, 6:44:55 PM


by: pfdietz

Says the proof has been formalized.

11/20/2026, 3:40:20 PM


by: chaosconst

I am trying to reproduce the lean on a 192-core aws machine, live on: <a href="https:&#x2F;&#x2F;cloud.infero.net&#x2F;rh-en" rel="nofollow">https:&#x2F;&#x2F;cloud.infero.net&#x2F;rh-en</a>.

11/20/2026, 5:41:12 PM


by: igravious

A new era for mankind has undoubtedly dawned.<p>All that is solid melts into air, all that is ineffable crystallizes.

11/20/2026, 5:54:28 PM


by: sp-ti-en-ma-in

Is this last normal year?

11/20/2026, 3:28:47 PM


by: mik1998

I feel sorrowful about this. Some sort of melancholy mixed with disappointment, and nostalgia. I suppose it was bound to happen eventually.<p>On another note, I always thought the Riemann hypothesis was false, with a counterexample that was too large or too small to be enumerated.

11/20/2026, 4:16:53 PM


by: kravmaga999

That&#x27;s insane

11/20/2026, 3:49:55 PM


by: objektif

What is the significance of this for us pheasants?

11/20/2026, 5:13:41 PM


by: matt3210

Stock bumped a few points on this news it seems then slipped back to where it was as people took profits. Math slop must be priced at already.

11/20/2026, 5:24:53 PM


by: nickalaso

In light of this, my recent comment, with latentsea&#x27;s subcomment is actually fucking hilarious.<p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=50492141#50493545">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=50492141#50493545</a>

11/20/2026, 4:05:01 PM


by: andyjohnson0

This is currently #1 on the home page. And there&#x27;s apparently a 250,000 page Lean proof.<p>And #2 is Feynman&#x27;s saying that &quot;If you can’t explain something in simple terms, you don’t understand it&quot;.<p>So: is anyone going to be able to explain <i>or</i> understand this? Or is that the old paradigm now?

11/20/2026, 3:27:23 PM


by: TaupeRanger

Very cool. Now when do we get the breakthroughs in real science and medicine?

11/20/2026, 6:15:24 PM


by: jmyeet

If you don&#x27;t happen to be a Math PhD but want a more detailed description of this problem that&#x27;s closer to the math than the average popsci Youtube video that makes weird analogies, this is about the best I&#x27;ve found [1].<p>[1]: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=yCZgKep5iBc" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=yCZgKep5iBc</a>

11/20/2026, 4:43:07 PM


by: mdberkey

I was here!

11/20/2026, 4:06:10 PM


by: adverbly

I was here

11/20/2026, 4:17:11 PM


by: CharlieItanium

Reminder, this &quot;proof&quot; is a quarter of a million pages. Every person singing praises of AGI here has yet to even read the small debrief above, much less actually down a single part of the proof.

11/20/2026, 3:29:22 PM


by:

11/20/2026, 4:29:02 PM


by: machina_ex_deus

The real danger is just around the corner, and it has nothing to do with superintelligence: if these models are so good at math, they might break modern cryptography. All it takes for massive worldwide chaos is these models solving elliptic curve cryptography. Superintelligence or not, if these models can solve cryptographically relevant math questions, faster than we can invent new cryptographic protocols, everything we rely on digitally can break. Imagine the internet without encryption. That includes credit cards, operating systems secure updates in a time where models find vulnerabilities every day, etc. I would start pushing post-quantum cryptography protocols not because of quantum computers, but because potentially any day now a model will come and break everything, and we need to have backup.

11/20/2026, 3:28:21 PM


by: pouetpouetpoue

what is the point of a lean proof? it seems to have the same level of interest than &quot;running&quot; a marathon with a moped and pretend that a record has been broken.<p>does this proof brings interesting new subjects and understandings or not?

11/20/2026, 5:11:07 PM


by: kittikitti

One of my recent, fully human, mathematical discoveries relies on the Reimann Hypothesis. It&#x27;s likely to still get dismissed because I don&#x27;t have a gatekeeper backing me up, but I can rest easier knowing this. I know I haven&#x27;t been nice to Anthropic but I truly appreciate this. Now all the haters will have to find something else to trivialize about my work!

11/20/2026, 3:17:20 PM


by: amir734jj

If Reimann hypothesis is true, then Goldbach conjecture is true as well. It&#x27;s major.

11/20/2026, 3:17:12 PM


by: ogogmad

GRH is a juggernaut. Nay, a keystone. Now number theory can get tidier and more user-friendly.

11/20/2026, 3:16:42 PM


by: simianwords

It’s funny that this happened after the IPO and not before. I’m sure he timed it that way

11/20/2026, 2:56:21 PM


by: expectfun

It&#x27;s wise to wait till it will be checked independently and let&#x27;s not just rush to conclusions yet,<p>BUT HOLY FUCK

11/20/2026, 4:10:27 PM


by: themarinho

its the new era

11/20/2026, 4:27:18 PM


by: makmanalp

Am I the only one who immediately thought, release the transcripts? I&#x27;d love to see how this came about.

11/20/2026, 3:43:46 PM


by: JumpCrisscross

&gt; I had a stream-of-consciousness jaunt<p>What? Why? Did nobody—not even an agent—copy edit this? Why is the first half a Turgenev novel?

11/20/2026, 3:08:39 PM


by: darkamaul

&gt; Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line.<p>I quite love the remark here, this feels quite poetic. And it shows, what most experts have already said, that the correct way to use AI to do great things is the Centaur model: frontier AI + domain expert.

11/20/2026, 2:48:30 PM


by: reworkal

&gt; No, Claude flew, and danced, and sparkled near that finish line.<p>Side note, anthropomorphizing your stochastic parrot sort of distracts me from your argument. Like sure guy might be a genius but as someone who knows nothing about you, reading this is very off-putting.<p>This is like my own opinion.. not saying what he wrote is bad, hope its real TBH

11/20/2026, 2:57:13 PM


by: whimsicalism

crazy times

11/20/2026, 3:49:36 PM


by: MadrasThorn

&gt;It was my secret in my student days that my plans were for this problem in decades, which meant then that I tried to become familiar with all tools that I could possibly recognize in the problem when it was time. Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line<p>yeah this is just another advertisement

11/20/2026, 6:43:37 PM


by: varispeed

That is great, but why Anthropic and other companies are allowed to hoard and gatekeep resources? Businesses and consumers already struggle with buying computing at prices they can afford.<p>To me this reads like: &quot;I bought all your water and solved my thirst, peasants!&quot;

11/20/2026, 3:03:30 PM


by: roland_nilsson

The (alleged) proof is 15 million lines of code I hear. Meanwhile, the #2 item on the HN frontpage right now is &quot;If you can&#x27;t explain something in simple terms, you don&#x27;t understand it&quot;.

11/20/2026, 4:21:47 PM


by: trotro

This is honestly not that surprising. Quasi-Riemann Hypothesis was apparently thought to be totally unreachable by number theory experts before the proof dropped by OpenAI in October. So I think we are at a point where mathematicians&#x27; opinion on whether a conjecture is easy or hard doesn&#x27;t really say much about the frontier models&#x27; ability to solve it.<p>Regardless, this is a historic moment for science and humanity in general. We still have 6 weeks to go, but I think 2026 will be seen as a pivotal year in history. The year when humans stopped being the smartest species on Earth.<p>It&#x27;s hard not to feel excited about the future and, obviously, a little bit scared.

11/20/2026, 3:30:11 PM


by:

11/20/2026, 2:38:52 PM


by: arduanika

For years, I&#x27;ve held RH in mind as my personal benchmark for... something. AGI, the Singularity, or whatever.<p>We don&#x27;t know yet whether this proof will hold, or how bad this stock market rout will be. But it would be a supremely funny ending to this whole story, for me at least, if they spent all these years and cap ex to build AIs, they IPO Anthropic, they prove Riemann, and then in that moment the whole bubble bursts and the hype collapses.

11/20/2026, 4:19:54 PM


by: science_markets

[flagged]

11/20/2026, 5:38:31 PM


by:

11/20/2026, 3:44:04 PM


by: klas_segeljakt

[dead]

11/20/2026, 7:17:03 PM


by: klas_segeljakt

[dead]

11/20/2026, 6:55:12 PM


by: dinkm

[dead]

11/20/2026, 2:48:35 PM


by: zengyu199

[flagged]

11/20/2026, 4:02:39 PM


by:

11/20/2026, 5:38:49 PM


by: hindsight9000

[dead]

11/20/2026, 3:04:39 PM


by: itsthecourier

[dead]

11/20/2026, 4:11:32 PM


by: faangguyindia

Imagine being an employee at one of the largest AI companies. You can do this:<p>grep -ri &quot;Riemann Hypothesis&quot; &#x2F;all-users-chats<p>Then let an LLM sift through everything, identify potential clues, and pick out promising candidates.<p>Just think about it: how many people use such LLMs and talk to them about their proofs or research? or their moonshot ideas?<p>You can have a super-high IQ, but you may not find anything worthwhile, yet a guy less intelligent than you may discover something extraordinary.<p>To capture that, you need a numbers game. One among billions of dumb people has a higher likelihood than one among 1,000 geniuses.<p>And LLM companies have all geniuses and ordinary people contributing research for free.<p>It would be foolish to assume, that people working for LLM companies are coming up with proof all by themselves.

11/20/2026, 2:45:22 PM


by: saberience

And what?<p>Is anyone’s life going to change, be made better?<p>Does anyone actually understand the proof? Is it possible to even understand the 100k pages of proof?<p>This is just useless noise to try and juice a company’s stock price.

11/20/2026, 3:52:20 PM


by: Noaidi

I also find the timing highly convenient. It’s not being a conspiracy theorist to want to check for any new bank transfers.

11/20/2026, 4:15:49 PM


by: gxd

[flagged]

11/20/2026, 2:48:28 PM


by: EcommerceFlow

[flagged]

11/20/2026, 2:45:06 PM


by: pussylover69

[flagged]

11/20/2026, 2:27:28 PM


by: blondie9x

[flagged]

11/20/2026, 2:49:15 PM