From what historians of science tell us, Gödel was in any case an absolute perfectionist who mostly published work that was bulletproof. His enormous body of notes surely contained some very sharp insights, but it is unlikely that he would have produced illuminating, conclusive new theorems by chance and without realizing it. He was a very harsh reviewer of his own work.
Maybe it's likely that he didn't have proofs, but surely had all kinds of theorems with at the very least some intuition of a proof. Recall that he wouldn't have had the same huge breadth and depth of mathematical machinery available to him that we have now.
The stuff about Gabelsberger shorthand sounds ludicrous to me. My father would write notes in a custom version of it until his death 2020.
Officially (i.e. in schools teaching stenography, back in the day) it was long ago replaced by "unified German shorthand" (Deutsche Einheitskurzschrift IIRC), but he always considered Gabelsberger to be superior.
Well, maybe he too was one of the 6 people in the world who could read it (or yes, maybe that was hyperbole, I'm not sure) :D
> Von Plato said he had started his Gödel project in 2017 because he was facing mandatory retirement from Helsinki University. “What should I do to make myself indispensable? Oh yes, I’ll start reading Gödel’s enigmatic shorthand pages!” He received a €2.5m (£2.1m) grant from the European Research Council, taught himself Gabelsberger shorthand – “I am one of only six people who can read it in the world” – and assembled a small team.
The notebooks were transcribed to regular German by a couple[0] who learned Gabelsberger just for that purpose. There is a bunch of analysis of them in Gödel's 5-volume Collected Works, iirc mostly volume 4. I'm surprised if there was any interesting mathematical content left to be found. The question that mathematicians wondered about the most was whether Gödel made any progress on the independence of the continuum hypothesis and axiom of choice (both later proved by Cohen). It appeared that he had some thoughts about it but didn't get anywhere. The Guardian article says Gödel was close to a proof of AC's independence but I'm skeptical.
The question the public wondered about the most was the loophole Gödel claimed to have found in the US Constitution, but they didn't find anything about that.[1]
There's a cool suspense novel "The God Proof" about finding a lost Gödel notebook extending Gödel's ontological "proof" of the existence of God.[2]
0. John Dawson (a logician) and his wife whose name I think is Cheryl (not sure). John Dawson later wrote the Gödel biography "Logical Dilemmas".
I can't remember the exact quote but I believe I read that Gödel confided to Einstein (in their walks in Princeton) on his deep worry of his work being forgotten. If only he knew.
Have you seen the manuscripts? Take a look at slide 10 of https://aitp-conference.org/2026/slides/JvP.pdf [1]. On slide 12 you can at least make out a few words (even before reading the English translation I could clearly read "Jede" and "Systems ist berechenbar"), but any part of the shorthand is completely unrecognizable. Also take a look at slide 6 that shows a bunch of word transcriptions in the shorthand. If I understood the article correctly, you have to intimately know the math involved to have any chance of decoding the content, as there is a lot of interpretation involved.
If one could wish for a best case publishing, I would like to have the page scans side by side with the German reading side by side with the English translation with running commentary by the research group.
a computer can't see the letter "c" or any letter. You train it based on data and pixels otherwise it is "completely un-recongizable." What is so special about these squiggles? That it is a shorthand? Are words in other languages not "completely unrecognizable" to you?
How would you happen write a comment that is so utterly similar to your [dead] sibling comment?
> to a computer the letter "c" is completely unrecognizable no concept of the letter or the shape etc. If you train an ocr algorithm to detect pixels in the shape of "c" you can teach the computer to read the letter. what is so complicated about these squiggles that you think a computer couldn't be trained to read them? Have you seen written text in other languages? It looks "completely unrecognizable"
It feels too similar to even be AI generated. Or is the temperature set way down?
Also, could the people that for whatever reason cannot be stopped from running commenting agents (it's against HN policy btw.) please at least use models that aren't dumb? Yes, it is because it is shorthand. At least read the sources I'm quoting, dear agent.
That would be the future of crowdsourcing - by submitting the same prompt as 1000 other people, you're basically donating your tokens to the cause of maybe spotting a valuable insight the others didn't get because of random output sampling :)
This is incredibly exciting! Gödel is unquestionably the most important logician in history, but unfortunately he published very little. For instance, the Dialectica interpretation was published approximately 20 years after his first ideas on the subject.
Plus, Jan von Plato is the world's leading expert on Gödel -- the number of people who can read Gödel's handwriting can be counted on the fingers of one hand, and I think that every single one is part of Jan von Plato's research group. If he is excited about the output, this is a huge deal.
The problem with Linear A is that we don't have a reference point like the Rosetta Stone to anchor it to other languages and scripts. It also has a small corpus, making it hard to find patterns in the texts.
I wonder how much expertise was actually needed to translate these notebooks. In other words, could a general purpose model one-shot these translations, or did we really need a subject matter expert to do the work? BTW - I'm not saying having a subject matter expert tackle the problem with Claude isn't useful - just wondering what the limits are!
This might be an actual concrete example of someone making themselves obsolete with AI.
> Von Plato said he had started his Gödel project in 2017 because he was facing mandatory retirement from Helsinki University. “What should I do to make myself indispensable? Oh yes, I’ll start reading Gödel’s enigmatic shorthand pages!” He received a €2.5m (£2.1m) grant from the European Research Council, taught himself Gabelsberger shorthand – “I am one of only six people who can read it in the world” – and assembled a small team.
I think you're missing my point. It's that once the workbooks are completely interpreted, which is happening at an accelerated pace thanks to AI, von Plato's value as an interpreter of the short hand becomes severely reduced, presumably as does his ability to get future grants.
I'd think there would be more competition to lead studies and receive grants now that the workbooks are accessible to a wider audience of logicians
If he was facing mandatory retirement nearly ten years ago it’s likely he might be ready for voluntary retirement, particularly after achieving this fantastic result.
The professor's value was/is in being one of a few people with the capability of interpreting the shorthand. By translating it with AI, his value as an interpreter begins evaporating as the translated workbooks become accessible to more logicians.
I actually got a very different impression. He could publish this interesting result, exactly because he had been working on this project i.e., had an interesting dataset to feed into Claude and had the expertise to interpret its output.
Compare him to those who are burning tokens to generate cringeworthy AI slop on social media. The value of a tool depends on the skill of the user.
It's clear though. Even if an AI had connected all the dots on the obsolete shorthand, it would never have recognized the human value. I doubt that any AI process would choose to spend the energy on interpreting this particular obsolete handwriting and then understand that it was from the hand of a genius.
Using AI to do your job does not mean that AI suddenly walks in the door and escorts you out and starts collecting your paycheck.
Someone is still driving these studies, and probably will be for a while. Or maybe not. But still we can reward this guy for opening up this can of theorems that had before.
Silliest case:
Imagine a dusty basement terminal screen open to claude code with "There we go! Cancer is cured as you requested, let me know and I'll send off the results" and nobody is around to approve it.
This extra detail makes the story even sadder, imo:
> Following Adele's admission to hospital after suffering a stroke in late 1977, Gödel refused to eat in her absence; he weighed 29 kilograms (65 lb) when he died of "malnutrition and inanition caused by personality disturbance" in Princeton Hospital on January 14, 1978
There was also a project to catalog and prove all the theorems in Ramanujan's notebooks. There's a book series by Bruce C. Berndt which covers the 5 notebooks:
"You don't get credit for the work you don't tell anybody about"
This has been a problem long before Godel: famously, Newton tried to get credit for inventing Calculus only after Leibniz published his work, leading to a bitter dispute that lasted long after they both died.
I wonder if this problem is now going to be significantly exacerbated by AI, as mathematicians are getting more incentives to keep their works secret —
— or, to the contrary, the community will get a push to publish incomplete ideas, scratch work, etc. early to establish a record of thinking about a problem before others did.
In the end, as always, it's a social problem created by " publish or perish " institutional incentives which reward (and require!) published results, but not the work that led to to them.
Maybe we'll see a solid societal use case for zero-knowledge proofs here, establishing one had a set of documents at a given date without revealing them.
Yes, you can. So maybe we'll see an adoption of that paradigm, or another. The value of a ZKP is that you can optionally prove some facts up front about the file, whereas a normal hash is all-or-nothing. You can't reveal part of the information because you need the entire contents to properly verify the hash.
At least that's how traditional Western universities used to work. The profs sometimes published something, and they taught students. Looking at science back then, it seemed to work out rather well.
I first came to know about Gödel's incompleteness theorem while watching a video by Veritasium (https://www.youtube.com/watch?v=HeQX2HjkcNo), it's around 30 minutes but a really good video. It's from the time before Veritasium became slop. The video talks about many things that are interconnected, like the halting problem, undecidable systems like the game of life, Hilbert's dream of a finite set of axioms to prove all mathematics, and, of course, a perfect explanation of Gödel's incompleteness theorems
Nice timing for the IPO. Cooked Nvidia numbers on the 17th are still missing, and the 19th will go well!
The AI was of course trained on existing pre-AI transcriptions of Goedel's shorthand, which he was taught in school around 1920 and which was widespread and only technically "obsolete".
In the first paragraph, we have an evidence-free insinuation that the foremost expert on this subject is some kind of shill running a campaign for Anthropic. What does this campaign involve? Is he just making it all up? Let me hear your theory of what is happening here and how it invalidates the claimed abilities of the model (or why the abilities do not matter).
In the second paragraph, we have what appears to be a claim that the model knows how to read Gödel's shorthand because it has been exposed to past examples of that shorthand. What is your point here? I think the model learning to generalize by training on a few examples is a point in its favor.
Why? His incompleteness work explains why software and math suck so he would already expect them to; they're rhetorical and can never fully be a closed loop body of knowledge as we can never with rhetoric concretely observe/define the far future or far past; merely simulate rhetorically
Ergo all defined axiomatic systems have sentences about the far past and future that cannot be defined in those languages with certainty. Most do not make the connection to physics. Gödel did and would 100% aware the limits exist and will always exist
Scaling laws aren't just for computers; humans cultures do not scale infinitely through time, just approximately 2-3 generations
As I understand immutable physical truth like this played a huge role in his early research and ultimately were the forcing function that led him to incompleteness
"all defined axiomatic systems have sentences about the far past and future that cannot be defined in those languages with certainty."
This is often misunderstood or misstated to mean that there are some things that are unknowable or unable to be solved. (I'm not saying that's what you were implying, I am referring to the common way I see his theorem being represented)
But what he really proved is that no single sufficiently powerful, consistent, effectively specified formal system can prove every arithmetic truth.
He didn't prove that there are mathematical truths that are unsolvable through any possible extension or alternative framework. It's just that one framework isn't enough to solve all problems.
You're just restating Gödel verbatim without considering what that would entail physically; they're implicitly unsolvable as it would take forever to solve them all; the universe would experience heat death before they're all solved.
Scaling issues do not just matter when making bridges and when hosting LLMs.
Back in Gödel's day physicists were like rockstars and all Gödel's output was 100% influenced by Einstein and the rest. Given the group's expertise in the domain the implicitly forever incomplete/never computed did not need to be spelled out explicitly.
"Unlocks" cool so Claude the trillion dollar technology did what any funded researcher with a computer vision algorithm could have done years ago. Wow so impressive.
The article mentions nothing about OCR and appears to state that transcription had already been done (by hand, OCR, or otherwise):
> On his return to Helsinki he began the process, showing Claude his transcriptions of the workbooks, as well as six books he had previously written on Gödel and other secondary scientific literature.
> “In a couple of days, I got to see a 650-page explanation of the workbooks. It had to be controlled page for page, but it’s basically a work that is impossible to do for human beings,” he said. “Humans read one page at a time and we tend to forget things. It takes us a year to read about 50 pages. We just are not able to keep in mind all the different parts.
> “The AI can connect all the pieces in a thousand pages. It can see all of Gödel’s writings as one piece.”
From what historians of science tell us, Gödel was in any case an absolute perfectionist who mostly published work that was bulletproof. His enormous body of notes surely contained some very sharp insights, but it is unlikely that he would have produced illuminating, conclusive new theorems by chance and without realizing it. He was a very harsh reviewer of his own work.
He did discover the functional interpretation in 1940 and publish it only in 1958, so it is highly probable that he has valuable unpublished results.
n=1
> highly probably
Ah, the optimism of today’s youth!
Probability is not a measure of frequency
Fair assessment but context matters. He is one of the greats, so it’s not as far fetched as it might sound.
Maybe it's likely that he didn't have proofs, but surely had all kinds of theorems with at the very least some intuition of a proof. Recall that he wouldn't have had the same huge breadth and depth of mathematical machinery available to him that we have now.
You could just read the article instead of speculating about what's in it. This isn't reddit.
Yes, it does state they found proofs and new theorems but also false ones too. I'm curious what new insights the ones that were proved hold.
It's really cool how AI is just able to decipher these and validate them with automated theorem provers. Crazy times!
The stuff about Gabelsberger shorthand sounds ludicrous to me. My father would write notes in a custom version of it until his death 2020.
Officially (i.e. in schools teaching stenography, back in the day) it was long ago replaced by "unified German shorthand" (Deutsche Einheitskurzschrift IIRC), but he always considered Gabelsberger to be superior.
Well, maybe he too was one of the 6 people in the world who could read it (or yes, maybe that was hyperbole, I'm not sure) :D
> Von Plato said he had started his Gödel project in 2017 because he was facing mandatory retirement from Helsinki University. “What should I do to make myself indispensable? Oh yes, I’ll start reading Gödel’s enigmatic shorthand pages!” He received a €2.5m (£2.1m) grant from the European Research Council, taught himself Gabelsberger shorthand – “I am one of only six people who can read it in the world” – and assembled a small team.
I think it's 6 people who can read Gabelsberger and know enough mathematical logic to make out what Gödel is saying.
The notebooks were transcribed to regular German by a couple[0] who learned Gabelsberger just for that purpose. There is a bunch of analysis of them in Gödel's 5-volume Collected Works, iirc mostly volume 4. I'm surprised if there was any interesting mathematical content left to be found. The question that mathematicians wondered about the most was whether Gödel made any progress on the independence of the continuum hypothesis and axiom of choice (both later proved by Cohen). It appeared that he had some thoughts about it but didn't get anywhere. The Guardian article says Gödel was close to a proof of AC's independence but I'm skeptical.
The question the public wondered about the most was the loophole Gödel claimed to have found in the US Constitution, but they didn't find anything about that.[1]
There's a cool suspense novel "The God Proof" about finding a lost Gödel notebook extending Gödel's ontological "proof" of the existence of God.[2]
0. John Dawson (a logician) and his wife whose name I think is Cheryl (not sure). John Dawson later wrote the Gödel biography "Logical Dilemmas".
1. https://en.wikipedia.org/wiki/G%C3%B6del%27s_loophole
2. https://jeffreykegler.github.io/personal/gp/God_Proof.pdf
Trump has been showing the loophole in real time. He would have shown more if the Democrats hadn't won.
I read it as knowing it at all but maybe it’s the mathematical shorthand subset.
I can't remember the exact quote but I believe I read that Gödel confided to Einstein (in their walks in Princeton) on his deep worry of his work being forgotten. If only he knew.
Now in 2026 every store use facial recognition to make sure you are ALWAYS remembered.
For short geological time horizons at least.
Anxiety is a great motivator
AI did not comment on the consistency or completeness of the notebooks.
True! but unprovable.
Funny thing is your comment is consistent and complete.
This is possible for things that have less power than arithmetic!
It'd be interesting to hear about these new 14 themes of logic Gödel pursued!
Article also mentions his notes were a few months work away from proving independence of CH decades before Cohen's result in 1963.
Yeah, these notes need to be just openly published. Then I could ask my own Claude for an explanation. (Not /s)
Or just share the Claude sessions as Tao did his.
You could likely "unlock" the manuscript yourself with a competent computer vision algorithm. It's nothing that fancy.
Have you seen the manuscripts? Take a look at slide 10 of https://aitp-conference.org/2026/slides/JvP.pdf [1]. On slide 12 you can at least make out a few words (even before reading the English translation I could clearly read "Jede" and "Systems ist berechenbar"), but any part of the shorthand is completely unrecognizable. Also take a look at slide 6 that shows a bunch of word transcriptions in the shorthand. If I understood the article correctly, you have to intimately know the math involved to have any chance of decoding the content, as there is a lot of interpretation involved.
If one could wish for a best case publishing, I would like to have the page scans side by side with the German reading side by side with the English translation with running commentary by the research group.
[1] via https://news.ycombinator.com/item?id=50487434 , thanks empath75
a computer can't see the letter "c" or any letter. You train it based on data and pixels otherwise it is "completely un-recongizable." What is so special about these squiggles? That it is a shorthand? Are words in other languages not "completely unrecognizable" to you?
What the…
How would you happen write a comment that is so utterly similar to your [dead] sibling comment?
> to a computer the letter "c" is completely unrecognizable no concept of the letter or the shape etc. If you train an ocr algorithm to detect pixels in the shape of "c" you can teach the computer to read the letter. what is so complicated about these squiggles that you think a computer couldn't be trained to read them? Have you seen written text in other languages? It looks "completely unrecognizable"
It feels too similar to even be AI generated. Or is the temperature set way down?
Also, could the people that for whatever reason cannot be stopped from running commenting agents (it's against HN policy btw.) please at least use models that aren't dumb? Yes, it is because it is shorthand. At least read the sources I'm quoting, dear agent.
That would be the future of crowdsourcing - by submitting the same prompt as 1000 other people, you're basically donating your tokens to the cause of maybe spotting a valuable insight the others didn't get because of random output sampling :)
This is incredibly exciting! Gödel is unquestionably the most important logician in history, but unfortunately he published very little. For instance, the Dialectica interpretation was published approximately 20 years after his first ideas on the subject.
Plus, Jan von Plato is the world's leading expert on Gödel -- the number of people who can read Gödel's handwriting can be counted on the fingers of one hand, and I think that every single one is part of Jan von Plato's research group. If he is excited about the output, this is a huge deal.
Very much need a statement in there as to if AI was used in the making of the manifesto, goes to the heart of the ethical soul of the manifesto.
I’d suggest you bake it into that header line on writtten by; something like this (but edited depending how ai was used):
“Written without AI assistance by the speakers and attendees of AIE Code Summit 2026 ”
Was this comment meant to be on the other neighbouring link (No AI Slop: A Manifesto - https://news.ycombinator.com/item?id=50488769)?
https://aitp-conference.org/2026/slides/JvP.pdf this seems to be slide deck with some of the results.
And here is the site of the research group involved. https://www.helsinki.fi/en/researchgroups/godel-enigma
Thank you! Godel's work get us comfortably uncomfortable with unknowability but damn, to think there's more out there! I'm dying to know more.
Let's try the Voynich Manuscript next
I’m been trying to use various frontier models to decipher lost ancient languages (like Linear A).
I wonder what tactics this used, so that I can leverage in my own work.
The problem with Linear A is that we don't have a reference point like the Rosetta Stone to anchor it to other languages and scripts. It also has a small corpus, making it hard to find patterns in the texts.
I wonder how much expertise was actually needed to translate these notebooks. In other words, could a general purpose model one-shot these translations, or did we really need a subject matter expert to do the work? BTW - I'm not saying having a subject matter expert tackle the problem with Claude isn't useful - just wondering what the limits are!
Uh, a decades old properly trained OCR algorithm could probably "unlock" these manuscripts
Right you are - this was "trivial", and everyone is "stupid" - except for you
This might be an actual concrete example of someone making themselves obsolete with AI.
> Von Plato said he had started his Gödel project in 2017 because he was facing mandatory retirement from Helsinki University. “What should I do to make myself indispensable? Oh yes, I’ll start reading Gödel’s enigmatic shorthand pages!” He received a €2.5m (£2.1m) grant from the European Research Council, taught himself Gabelsberger shorthand – “I am one of only six people who can read it in the world” – and assembled a small team.
And via earthshaking victory, at that, wow.
Okay let me know when AI gets itself a 2.1M Euro grant from ERC as the PI.
I think you're missing my point. It's that once the workbooks are completely interpreted, which is happening at an accelerated pace thanks to AI, von Plato's value as an interpreter of the short hand becomes severely reduced, presumably as does his ability to get future grants.
I'd think there would be more competition to lead studies and receive grants now that the workbooks are accessible to a wider audience of logicians
If he was facing mandatory retirement nearly ten years ago it’s likely he might be ready for voluntary retirement, particularly after achieving this fantastic result.
Indeed, he is 75 years old.
I would his not be in verifying the AI is output?
It's literally an example of someone keeping themselves from becoming obsolete with AI.
The professor's value was/is in being one of a few people with the capability of interpreting the shorthand. By translating it with AI, his value as an interpreter begins evaporating as the translated workbooks become accessible to more logicians.
He might see his value in his role of carrying the knowledge itself rather than being its gatekeeper.
I actually got a very different impression. He could publish this interesting result, exactly because he had been working on this project i.e., had an interesting dataset to feed into Claude and had the expertise to interpret its output.
Compare him to those who are burning tokens to generate cringeworthy AI slop on social media. The value of a tool depends on the skill of the user.
And he didn't want to milk that for the rest of his life because he's a smart guy that has other skills and interests
That's what he thinks, surely.
It's clear though. Even if an AI had connected all the dots on the obsolete shorthand, it would never have recognized the human value. I doubt that any AI process would choose to spend the energy on interpreting this particular obsolete handwriting and then understand that it was from the hand of a genius.
But is he needed once he finished writing the prompt?
Using AI to do your job does not mean that AI suddenly walks in the door and escorts you out and starts collecting your paycheck.
Someone is still driving these studies, and probably will be for a while. Or maybe not. But still we can reward this guy for opening up this can of theorems that had before.
Silliest case: Imagine a dusty basement terminal screen open to claude code with "There we go! Cancer is cured as you requested, let me know and I'll send off the results" and nobody is around to approve it.
Maybe the notebooks will contain details of his "loophole", or maybe it will have been proven true in real life by then.
> Gödel himself, when rereading his notes, would occasionally add a marginal remark: falsch! [incorrect] or schlecht! [bad].”
This is probably my favourite part of the article. On brand.
> Increasingly paranoid in his later years, he refused to eat food that was not cooked by his wife, and died of starvation in 1978
genius often comes with side effects
This extra detail makes the story even sadder, imo:
> Following Adele's admission to hospital after suffering a stroke in late 1977, Gödel refused to eat in her absence; he weighed 29 kilograms (65 lb) when he died of "malnutrition and inanition caused by personality disturbance" in Princeton Hospital on January 14, 1978
https://en.wikipedia.org/wiki/Kurt_Gödel#Personal_life_and_d...
My favourite end-of-life weirdo has to be Grothendieck.
There is a similar project to digitize and analyze Grothendieck's notes and letters.
There was also a project to catalog and prove all the theorems in Ramanujan's notebooks. There's a book series by Bruce C. Berndt which covers the 5 notebooks:
https://www.amazon.com/stores/Bruce-C.-Berndt/author/B001HN3...
It would be interesting to point Fable or Argon at the notebooks and see if they can find some interesting new material in or based on this work.
The line between Genius and Crazy is but hairs width
"You don't get credit for the work you don't tell anybody about"
This has been a problem long before Godel: famously, Newton tried to get credit for inventing Calculus only after Leibniz published his work, leading to a bitter dispute that lasted long after they both died.
I wonder if this problem is now going to be significantly exacerbated by AI, as mathematicians are getting more incentives to keep their works secret —
— or, to the contrary, the community will get a push to publish incomplete ideas, scratch work, etc. early to establish a record of thinking about a problem before others did.
In the end, as always, it's a social problem created by " publish or perish " institutional incentives which reward (and require!) published results, but not the work that led to to them.
This, I hope, will finally change.
Maybe we'll see a solid societal use case for zero-knowledge proofs here, establishing one had a set of documents at a given date without revealing them.
you can just write a doc with your proof/idea, hash it and post that hash on twitter
Yes, you can. So maybe we'll see an adoption of that paradigm, or another. The value of a ZKP is that you can optionally prove some facts up front about the file, whereas a normal hash is all-or-nothing. You can't reveal part of the information because you need the entire contents to properly verify the hash.
ceiiinosssttuv
Thanks for putting me onto this, that was new to me. And of course Huygens was all over this idea.
so what would the new incentives be? don't publish... and get paid anyway?
At least that's how traditional Western universities used to work. The profs sometimes published something, and they taught students. Looking at science back then, it seemed to work out rather well.
True.
If possible, this would be ideal.
I first came to know about Gödel's incompleteness theorem while watching a video by Veritasium (https://www.youtube.com/watch?v=HeQX2HjkcNo), it's around 30 minutes but a really good video. It's from the time before Veritasium became slop. The video talks about many things that are interconnected, like the halting problem, undecidable systems like the game of life, Hilbert's dream of a finite set of axioms to prove all mathematics, and, of course, a perfect explanation of Gödel's incompleteness theorems
A little disappointed to learn it was a guy prompting and not an OpenAI agent swarm reward hacking the goal of "prove as many theorems as possible".
typical newspaper blown up wording.
I hear they're all incomplete
Nice timing for the IPO. Cooked Nvidia numbers on the 17th are still missing, and the 19th will go well!
The AI was of course trained on existing pre-AI transcriptions of Goedel's shorthand, which he was taught in school around 1920 and which was widespread and only technically "obsolete".
https://link.springer.com/book/10.1007/978-3-030-67606-3
In the first paragraph, we have an evidence-free insinuation that the foremost expert on this subject is some kind of shill running a campaign for Anthropic. What does this campaign involve? Is he just making it all up? Let me hear your theory of what is happening here and how it invalidates the claimed abilities of the model (or why the abilities do not matter).
In the second paragraph, we have what appears to be a claim that the model knows how to read Gödel's shorthand because it has been exposed to past examples of that shorthand. What is your point here? I think the model learning to generalize by training on a few examples is a point in its favor.
I think we'll find godel's most pungent critiques of modern logic & mathmatics to be the maddening limits of LLMs.
Why? His incompleteness work explains why software and math suck so he would already expect them to; they're rhetorical and can never fully be a closed loop body of knowledge as we can never with rhetoric concretely observe/define the far future or far past; merely simulate rhetorically
Ergo all defined axiomatic systems have sentences about the far past and future that cannot be defined in those languages with certainty. Most do not make the connection to physics. Gödel did and would 100% aware the limits exist and will always exist
Scaling laws aren't just for computers; humans cultures do not scale infinitely through time, just approximately 2-3 generations
As I understand immutable physical truth like this played a huge role in his early research and ultimately were the forcing function that led him to incompleteness
"all defined axiomatic systems have sentences about the far past and future that cannot be defined in those languages with certainty."
This is often misunderstood or misstated to mean that there are some things that are unknowable or unable to be solved. (I'm not saying that's what you were implying, I am referring to the common way I see his theorem being represented)
But what he really proved is that no single sufficiently powerful, consistent, effectively specified formal system can prove every arithmetic truth.
He didn't prove that there are mathematical truths that are unsolvable through any possible extension or alternative framework. It's just that one framework isn't enough to solve all problems.
You're just restating Gödel verbatim without considering what that would entail physically; they're implicitly unsolvable as it would take forever to solve them all; the universe would experience heat death before they're all solved.
Scaling issues do not just matter when making bridges and when hosting LLMs.
Back in Gödel's day physicists were like rockstars and all Gödel's output was 100% influenced by Einstein and the rest. Given the group's expertise in the domain the implicitly forever incomplete/never computed did not need to be spelled out explicitly.
"Unlocks" cool so Claude the trillion dollar technology did what any funded researcher with a computer vision algorithm could have done years ago. Wow so impressive.
Why didn't they do it then?
This kind of comment is ignorant and damaging to HN.
Do better.
AI sycophancy is ignorant and damaging to HN. So is misattributing decades of OCR research to the company that stole everyone's data.
The article mentions nothing about OCR and appears to state that transcription had already been done (by hand, OCR, or otherwise):
> On his return to Helsinki he began the process, showing Claude his transcriptions of the workbooks, as well as six books he had previously written on Gödel and other secondary scientific literature.
> “In a couple of days, I got to see a 650-page explanation of the workbooks. It had to be controlled page for page, but it’s basically a work that is impossible to do for human beings,” he said. “Humans read one page at a time and we tend to forget things. It takes us a year to read about 50 pages. We just are not able to keep in mind all the different parts.
> “The AI can connect all the pieces in a thousand pages. It can see all of Gödel’s writings as one piece.”