How dare you use variables in your math!


Tucker Carlson interviewed some random guy who builds kilns for a living — that could be interesting, except it turns out this Jess Elofson guy doesn’t talk about art or potter, but instead whines about algebra. He doesn’t get it. He thinks it’s a stupid, nonsensical procedure where you pointlessly mix letters with numbers. He’s an idiot, and so is Tucker.

“Why does algebra put letters into numbers?” Carlson asked.

“I think just to f**k with you,” Elofson said.

“So you deal with math in your job every day. You’ve already, I think, demonstrated your grasp of math is high”

“I mean, functional math. Yeah, yeah, for sure.”

“Well, if you can build a wood-fired kiln, you understand math,” Carlson said. “But you didn’t add letters to numbers in the process of doing that?”

“That’s the weirdest thing,” Elofson replied. “Nowhere did I find any letters in my math to calculate the size and structural integrity of a catenary arch. Isn’t that weird? Oh, that was weird. Because every math teacher was always like, you need to know some algebra. I’ve yet to encounter a single person in my life that uses algebra.”

“Actually?” Carlson asked.

“Have you?”

“Yeah, but I mean, I live in a fake world,” the former Fox News host said.

“That’s why you use fake math with letters. Congratulations to everybody out there in the matrix. Good job. You’re good little goy slaves,” Elofson said.

“That’s amazing.”

I use algebra automatically all the time, and even teach people how to use it. You can’t do physics or chemistry or biology or cooking or art or construction without it. How does he calculate how much brick he needs given a certain size of kiln he’s building, to use a simple example?

I don’t think Elofson’s grasp of math is at all high. He wouldn’t meet the minimal requirements for admission to my university. But then, I live in the real, physical world that requires some understanding of relationships and numbers and proportions and ratios, unlike these two guys who can get by on elementary anti-semitism.

American education is failing, given the example of these two gomers.

Comments

  1. says

    Sounds to me like Carlson is trying to cash in on the kook side of Joe Rogan’s audience. This guy sounds like exactly the kind of guy who Rogan would breathlessly urge on.

  2. timothyeisele says

    Elofson clearly doesn’t understand what algebra is, or when he is doing it. If he’s doing any structural calculations at all, he is doing algebra. He just doesn’t realize it, because he is filling in his variables with actual numbers on the fly, without understanding that they were variables before he did that.

  3. raven says

    “Well, if you can build a wood-fired kiln, you understand math,” Carlson said.

    That is wrong.

    We humans have been making pottery for 20,000 years all over the world and using wood fired kilns.
    Long before numbers and algebra were invented.

    Good job. You’re good little goy slaves,” Elofson said.

    Using “gay” as an insult.
    The transcript says “goy” but that doesn’t make any sense. The word was probably “gay”.

    These guys are just babbling idiots pretending that our modern technological civilization wasn’t invented and isn’t run by people far more educated and intelligent than they are.

  4. timothyeisele says

    @raven: “goy” makes sense if he is trying to insinuate that “The Jews” rule the world and that all the non-jewish “goys” are their slaves.

  5. raven says

    @raven: “goy” makes sense if he is trying to insinuate that “The Jews” rule the world and that all the non-jewish “goys” are their slaves.

    Oh. OK.

    That makes him sound even dumber though.
    The Jews don’t rule the world and if I was a goy slave, I’d know it.

  6. raven says

    @raven, is our modern technological civilization really “run by people far more educated and intelligent than they are?”

    Yeah, good point.

    Our technological civilization was created and is maintained by people far more educated and intelligent that they are.

    These days, parts of it are lead by idiots that are their intellectual equals though.
    I’m sure Trump knows as much algebra as Elofson and Tucker.

  7. robro says

    Maybe Fucker Carlson is launching his comedy career? He did the math and realized that all that serious scowling, grumbling, and grousing is hurting his view numbers. And it doesn’t take a math genius using algebra or even basic accounting math to know that means his pocket book. So maybe a little levity will get him more eyeballs, and lift those view numbers. Et voila, more money.

    Plus, the America people are on the outs with his favorite Scowler-in-Chief, particularly those Americans in Canada and Mexico. Anyway, Donny is almost done…as in his goose is cooked. It’s been over 25 years since Donny first tried his scowling president act. That’s about the attention span of the American public.

    Just kidding.

  8. drmarcushill says

    Raven @6 “The Jews don’t rule the world and if I was a goy slave, I’d know it.”

    Ah, but that’s the cleverest part of the whole thing. They rule the world but manage to hide the fact that the goyyim are all their slaves. Probably using the space lasers. Wake up, sheeple!

  9. christoph says

    It really floored me in high school when I found out there was a practical application for algebra. I thought they were just teaching it for the sheer sadistic pleasure.

  10. submoron says

    P Z, this is Nobby and Fred Colon in Jingo surely!
    “There’s al-gebra. That’s like sums with letters. For…people whose brains aren’t clever enough for numbers, see?

  11. says

    Yes, his arguments just don’t add up!
    And, I quote: “I live in a fake world,” the former Fox News host said.”
    Truer words were never spoken.

  12. robro says

    charley @ #14 — And for lots god-bothering goofballs: X / 0 = ∞

    Recursive Rabbit @ #12 — Per my own research (i.e. Wikipedia) the so-called “Arabic numerals” originated in Vedic India.

  13. Akira MacKenzie says

    I got a feeling this clown dropped out of high school in his freshman year and is damn proud of it.
    Either that, or he was homeschooled.

  14. Akira MacKenzie says

    One more and then I’ll move on:

    <

    blockquote>American education is failing, given the example of these two gomers.

    What do you expect in a country where most of the citizens are anti-intellectual and think education is a waste of their time? Unless we do something drastic, we’re heading back to the Dark Ages where only an elite had any education at all and even what they knew was sparce.

  15. drdrdrdrdralhazeneuler says

    New Rule: Right-wing instigators don’t get to use any of the technology that was built using variables.

    Welcome to the jungle.

  16. says

    Oh, just one more comment based on later ones I read:
    Yes, it’s a nasty thought, but, what about those of us that are gay goyim? Are they going to use those kilns as gas chambers.
    And, note that @22 Akira is also taking heed of what PZ admonishes. And Akira is sooo right: the war on education and intellectualism is rapidly pushing us down the death spiral into another dark ages!

  17. says

    Aren’t kiln-builders experts in heat transfer, thermodynamics and the Carnot cycle? This guy’s probably a mathematical genius, Dr. Myers.

  18. says

    Thinking about it, regular arithmetic is just solving for X where X is the only thing on one side of the equation. Algebra is an extension of that process to include multiple variables in messy situations. Imagine this guy also complains about “word problems” that expect you to sort out the numbers yourself and how to put them into an equation you can then crunch.

  19. says

    “I’ve yet to encounter a single person in my life that uses algebra.”

    There you have it. Not only is he an ignoramus, but that also describes his social circle. Virtually every adult I know, knows algebra. In contrast, these folks seem proud of their innumeracy. Are they also proud to be illiterate?

    And what is so hard about building a kiln? Building isn’t designing. It’s quite possible that he simply copies plans that other people designed, and is reasonably good at using tools.

  20. John Morales says

    I reckon he’s trolling.
    For sure, most people wouldn’t know a catenary, or be able to tell it from any other arch.

    (y = a \cosh\left(\frac{x}{a}\right) where a is the scaling factor, y is the height, x half the span (center to edge))

  21. CompulsoryAccount7746, Sky Captain says

    Rando:

    A LOT going in the description of the show and it’s sponsors. [Screenshot]

    [Title] The dark forces destroying art, beauty and robbing you of your God-given creativity (1h51m)

    The most honest people in any society are the artists, because they’re committed to beauty, which is truth. Jess Elofson is one of the few we have left. He’s a welder.

    Paid partnerships with:
    Ethos: Protect your family with life insurance […]

    Vulnerable People Project: [SaveWestBankChristians dot com]

    Two. Hours.
     
    UffDah! @21: Yes!
    InfoWars – Tradwife explains her lifestyle (9:27)

    traditional gender norms are so important to me personally, I live 18 hours a day inside of a crate when my husband is at work or in bed. […] I can’t tell you how many wives I know who got left out while their husbands were away and ended up scratching holes in the furniture or getting into the people food.

    * The cheese reference at 2:22.

  22. birgerjohansson says

    I found algebra seriously challenging, and I never got the hang of matices and vectors.
    But I realise those who master it can do useful work with it, just as ‘quantum’ stuff is useful.

  23. says

    The most honest people in any society are the artists, because they’re committed to beauty, which is truth.

    Which is why fascists are trying to replace artists with AI slop. And why wingnuts get offended every time a movie or book gets attention. They also have chatbots parroting vibe math through text prediction instead of actual statistical analysis.

  24. gijoel says

    Just in, President Trump has ordered the National Academy of Sciences to rename algebra to Americabra. Polls reveal that 70% of American are cringing at this Administration. That’s up by two points from a poll taken last week.

  25. mykroft says

    Like it or not, but American education (up through high school) is going to change a lot with the introduction of AI. Perhaps it will force teachers to go beyond rote memorization of a set of acts. Have students demonstrate that they understand the concepts, rather than training to take standardized tests. The good teachers do this, but all too many teach to the test. Ultimately, AI can help with education by providing personalized tutoring on subjects, but that won’t be common for a while.

  26. John Morales says

    [For those who might miss the allusiveness of the nym: the computer in the book is HOLMES IV (High-Optional, Logical, Multi-Evaluating Supervisor, Mark IV, the narrator (Manuel), gives it the nickname Mycroft Holmes (Mike for short) after Sherlock Holmes’ smarter brother — https://en.wikipedia.org/wiki/The_Moon_Is_a_Harsh_Mistress ]

    I do concur with #35. As in, ‘can help’, not ‘will replace’.

  27. indianajones says

    I mean, technically, you can do maths without numbers even. Just scratch however many strokes in the ground with a stick. Not only that but you can do it with the logic rules AND and NOT and not bother with addition, or calculus or anything else. Computers do it all the time. It just takes a while to scratch out Pi, for instance, on the ground is all.

  28. zetopan says

    “So you deal with math in your job every day. You’ve already, I think, demonstrated your grasp of math is high”

    It is no surprise that one innumerate fool would praise another innumerate fool. This sort of ignorant person promoting another ignorant person is completely commonplace. Years ago multiple scientific illiterates (predominately, conservatives) were quite busy promoting idiot Joseph Newman and his perpetual motion machine.

  29. John Morales says

    I mean, technically, you can do maths without numbers even. Just scratch however many strokes in the ground with a stick.

    Technically, those scratches encode numbers. Or dots. Or magnetic patterns you can’t see.

    Numbers are a concept, not a particular visual representation.

    (OTOH, one can indeed do geometry without numbers, e.g. compass & straightedge, so your claim holds)

  30. zetopan says

    timgueguen@1: “Sounds to me like Carlson is trying to cash in on the kook side of Joe Rogan’s audience. This guy sounds like exactly the kind of guy who Rogan would breathlessly urge on.”

    Absolutely, Rogan has promoted scientific illiterate, math illiterate, (both of which describe Rogan as well) and mentally ill, Terrence Howard a “jenius” for “proving” that one times one is actually two,.and “inventing” new form of math and physics he calls “Terryology”. It is loons promoting loons, all the way down.
    https://www.independent.co.uk/news/people/terrence-howard-thinks-1×1-2-has-a-secret-system-called-terryology-and-spends-17-hours-a-day-making-nameless-plastic-structures-10502365.html

  31. CompulsoryAccount7746, Sky Captain says

    raven @6: “The Jews don’t rule the world and if I was a goy slave, I’d know it.”

    Fascists Really Believe It

    I postulate seven key neofascist narratives. These are: [*snip*]

    in taking on male roles, white women have abandoned their natural one: to breed the white race. Remember, because overwhelmingly white-majority nations are the foundation of ‘Western civilisation’, and because the racially inferior will kill all the whites if given the chance, this is an existential problem. One that neofascists rail against endlessly.

    And in this we can see the functional purpose that antisemitism plays in fascism. Put simply: if white men are so superior, how have white racial nations allowed themselves to be tricked in such a way? The answer tends to be to imagine a racial-cultural nation that has the intelligence of whites—it would have to—but that hates the white nations and wishes to destroy them. And that’s the Jews.
    […]
    Imagine for a second you’re the hero of some supernatural action thriller movie, a Van Helsing fighting off vampires or werewolves. That’s how bought-in fascists see the world. What you’re doing is not just necessary, it’s awesome. You’re the hero for doing it! White liberals appear to them as the human vampire helpers, pathetically aiding the monsters, not realising we’ll ultimately be eaten.
    […]
    a representation of the world that is insane. Hundreds of millions could die because of ideas intellectually and morally equivalent to the ‘lizard people’ conspiracy theory. […] There is […] no mastermind who can be reasoned with […] The leaders are even more deranged than the followers.

  32. astringer says

    Genuine question for the community on this topic: I have been in awe of the 3rd-6th century arch at https://en.wikipedia.org/wiki/Taq_Kasra since reading Dahl’s autobiog. One can conceive of building this arch without “maths” by using a chain (…caterary) to make a model arch: ironically, shown here https://en.wikipedia.org/wiki/Catenary: scroll down to find “Catenary arch kiln under construction over temporary form”. Does anyone have older examples of these free-standing anthropogenic arches? Did they stumble on the shape? Or preceed Hooke with the maths, not just the method?

  33. Silentbob says

    I do not work in a STEM profession, but I use algebra often. Often I’ll find myself expressing a problem as an equation and solving for the unknown. I think it’s as basic as “readin’, writin’ and arithmetic”.

  34. EigenSprocketUK says

    Fun game to play with primary-age kids is any variant on the “think of a secret number… now multiply by four… add 8… divide by two. Now what have you got? Now I can tell you your secret number” game. Kids love it, and at the right point they really love finding out how it works.
    This guy can understand that game, but he excised that part of his brain to pwn the libz.

  35. robert79 says

    @ EigenSprocketUK

    Oddly enough, the only variant of that I remember from that age was: “think of a secret number, multiply by 4, add 8, divide by 2, add 4, divide by 2, subtract the original secret number, you’re now thinking of the number 6 aren’t you?”

    I also loved figuring out how that worked (and refiguring it just now, I’m pretty sure the example I remember was more complicated but the basic idea holds).

  36. robert79 says

    and of course I just made a math error writing out algebra in words, and that is why you need algebra…

  37. says

    You know how some stoner groups are depicted as having “profound” deepities while they’re high based in wordplay and grand speculation about the nature of the universe? That’s what Carlson and Elofson sound like to me. And they’re allegedly sober.

  38. John Morales says

    [related]

    https://www.abc.net.au/news/2026-09-11/racing-to-solve-maths-problems-pointless-tristan-buckmaster-says/107141628
    ↓
    Racing to solve maths problems now ‘pointless’, says Australian at centre of OpenAI controversy

    Artificial intelligence has made the race to publish mathematical breakthroughs pointless, according to the Australian professor at the centre of a bitter dispute over a claimed solution to one of the world’s most difficult unsolved mathematical problems.

    The New York University professor also warned Australia about the centralisation of resources into the hands of a small number of privately owned foreign companies, which were now publishing solutions to maths problems that had puzzled people for decades.

    […]

    For now, Professor Buckmaster said he wanted to finish existing work and support his PhD students and postdoctoral researchers, rather than chase another Millennium Prize.

    “I think it’s more important to re-evaluate what math is,” he said.

  39. CompulsoryAccount7746, Sky Captain says

    From the article @51:

    OpenAI claimed it had solved the Navier-Stokes problem, one of seven fiendishly difficult Millennium Prize problems that come with a US$1 million reward

    Loring Craymer (Retired DoD supercomputing): “One of the more eye-popping features of the OpenAI Navier-Stokes result is that it reportedly cost $23M of machine time. If true, that highlights the difference between community (academic research trains grad students) and corporation even further.”
     
    Tim Henke (Physicist):

    Gro-Tsen (Mathemetician): [OpenAI announcement 3 months ago solving “unit distance” problem with experts’ comments]
    these (OpenAI-chosen) experts seem to mostly agree the main reason humans were stuck on the problem is they were trying to prove it rather find a counterexample.
    […]
    But what I think is missing from this kind of announcement is that we don’t really know how many open problems OpenAI tried to get their LLM to solve, and in which domains. E.g., are they tackling Erdős problems because those are famous, or because LLMs are better at them?

    Tim Henke: if I were an AI company trying to create a perception that AIs were going to replace mathematicians, what I would do is grab a large set of high-profile open problems and just start grinding counterexamples until I’ve found one, and then only report what went into finding that specific one.

    Some open problems are bound to be false, and the counterexample simply hasn’t been found yet bc people think it’s true, and finding counterexamples is boring. So you *will* find one. When you do, you can say you’ve “solved” a big open problem and by only reporting that problem you’ve made it look easy.

    So now you can say you got an Annals-level paper for pretty low costs. Media will report it breathlessly as if you one-shot it and the Death of Mathematics will be the talk of the day. Frankly you might even be able to pull this off w/o AI at all if the counterexamples can be checked algorithmically.

    [Screenshot: Noam Brown (OpenAI) tweet on Aug 28, following an Aug 1 OpenAI announcement solving 10 problems]
    I kinda consider this a soft confession to what I was theorising.
    1) He admits they only care about maths in so far as it brings good PR.
    2) They’re only interested in announcing results that make it sound like they’re gonna replace mathematicians.
    ∴) They’re optimising for impressive-soundingness.

    Tim Henke:

    Rando: Do you think all of the 10 problems that were solved fit this criteria?

    I can’t say that with confidence, as I haven’t had time to investigate all of them, but several of them yes. Two problems were solved just by connecting 2 results in the literature, to the point some mathematicians called misconduct to even claim ownership of the result; others deserved the credit.

    Some others were counterexamples. We also know from prompts that their approach was creating armies of agents, each of which attempting a different strategy with a harness preventing them from converging. In the successful cases that I know of, this ended up producing an overlooked elementary proof.

    SciAm – OpenAI’s latest math breakthroughs commit research misconduct

    Tim Henke:

    Rando: To be fair (because I think we should, lest we show biases), very few academics advertise their paper rejections.

    Of course! But the average academic is also not claiming to completely overturn academia. We roughly know how the next academic is going to perform based on previous academics. But if you’re selling an image that future maths will be prompt engineering, it’s important how successful that really is.

  40. John Morales says

    CA7746, you are trying to say they are not very good at it.

    It must take a lot of work to try to ignore reality; no matter how feebly, they can evidently actually do it.
    Better than people, even.

    cf. https://ourworldindata.org/grapher/ai-frontiermath-over-time

    The bots are solving problems that have been open for tens, even hundreds of years.
    Hundreds of them, just this year.

    Loring Craymer (Retired DoD supercomputing): “One of the more eye-popping features of the OpenAI Navier-Stokes result is that it reportedly cost $23M of machine time. If true, that highlights the difference between community (academic research trains grad students) and corporation even further.”

    Does it matter? It did what had not been hitherto done by humans.
    Were you not one of those who told me the bubble had burst years ago, and that LLMs are useless?

    Bleeding-edge tech is always kinda expensive.
    But then, wanna check the power/cost/performance improvement of computers since their inception?

    (Up to 15 orders of magnitude; a decade hence, that $23M might well become $0.0023)

  41. CompulsoryAccount7746, Sky Captain says

    “Were you not one of those who told me the bubble had burst years ago”

    No, the opposite. I’ve linked stories of economy-threatening financial shenanigans.

    “you are trying to say they are not very good at it. […] they can evidently actually do it. Better than people, even.”

    Leiden Declaration on Artificial Intelligence and Mathematics

    Mathematicians have a choice about whether and how to adopt artificial intelligence in the conduct of their research. […] This Declaration calls upon mathematicians to exercise this responsibility, and provides recommendations for individuals, institutions, government, and industry.
    […]
    Current automated techniques can produce plausible but unreliable (or even incorrect) arguments which are difficult to distinguish from correct mathematical proofs. This applies not only to informal arguments, but also to formalizations, where the difficulty lies in the translation between computer-encoded and human presentations of concepts. These fast-moving developments put our present system of review under increasing pressure, jeopardizing our ability to implement traditional standards for the correctness, transparency, and independent verifiability of proof.
    […]
    Leiden University in the Netherlands hosted a conference entitled Mechanization and Mathematical Research. The around 60 participants from 10 countries comprised mathematicians, computer scientists, philosophers, historians and social scientists, including those with experience in industry and in government. During the eight months following the conference, a smaller working group developed this Declaration, with extensive feedback from the mathematical community. The Declaration reflects artificial intelligence technologies and mathematical practice as of May 2026. [3916 Signatories]

  42. thistledown says

    “You can’t do physics or chemistry or biology or cooking or art or construction without it. How does he calculate how much brick he needs given a certain size of kiln he’s building, to use a simple example?”

    That’s not altogether true, certainly not with regard to cooking, art, and construction. Modern algebra has been worked out only since the sixteenth century, and people were quite good in those fields millenia before that. But there are good reasons why we use algebra. No one — but no one — would want to talk about number theory the way Euclid and Diophantus did.

  43. John Morales says

    Ahem, thistledown.

    I myself like a bit of number theory the way Euclid and Diophantus did.
    I like the odd bit of mathy fun.

    Here:

  44. Silentbob says

    @ 55

    From Morales link:

    The degree of human intervention in AI-generated mathematical proofs may vary and remain unclear. Subjects responsible for publishing such proofs, including individuals and AI companies such as OpenAI, often fail to disclose used prompts, structure of pipelines used as a model workflow, or specific models used for generating proofs. The publication of such proofs may not satisfy standards of science articles. It may not have passed independent review or peer review, and instead be published on corporate websites, personal websites, or social media.

    I believe “<snicker>” is the time honoured response. X-D

  45. Silentbob says

    BTW Mano has a post on this topic. For those who don’t know Mano is a retired Professor of Physics and his blog is very good (just don’t expect the histrionics tolerated here).

    Rather than just mock, Mano describes his own epiphany as a boy when he went from not “getting” algebra, to seeing the point clearly.

    To paraphrase (you should read the original post), normally you solve a problem “forwards”, starting with what is known to arrive at what was unknown.

    His insight that “clicked” was that in algebra you start with the unknown to arrive at the known. So you have to go backwards, but you do so by manipulating the problem to go forwards.

    Simple example:
    forwards
    6 multiplied by 9 = ?
    54
    backwards
    X multiplied by 9 = 54
    make it forwards by dividing both sides by 9 and reversing
    54/9 = X = 6

    I thought this a very concise and accurate way of conveying the concept to people who are like, “I get numbers but what is algebra”? You can tell Mano was an educator. X-D

    https://freethoughtblogs.com/singham/2026/09/13/what-the-hell-is-x/

  46. KG says

    I think it’s clear AI is good at certain aspects of mathematics – basically, those that are amenable to something close to “brute force” search through a vast range of possibilities – finding counterexamples to conjectures, or trying whether putting together two, three.. several existing findings in all possible ways can get you somewhere. And mathematicians can and do use it for that purpose – to do the boring bits, basically. But talking to a friend who is a professor of pure maths, he says the important advances are generally about discovering/inventing* new kinds of mathematical objects or relations, and so far, AI has not shown any ability to do that. I haven’t seen him since the Navier-Stokes result was announced, but (1) Again, it was the finding of a counterexample, and (2) In this case, even if the accusations of theft of results – of a near-complete proof – turn out to be wrong (and it’s not clear we’ll ever know), it is clear the prompts used could not have been devised without the prior work of mathematicians who had found counterexamples to the Euler equations – which correspond to the Navier-Stokes equations under the unrealistic conditions of zero viscosity and thermal conductivity. In short, this was a problem which was widely expected to be solved soon. Now if AI solves P=?NP or proves the Riemann hypothesis** in the immediate future, it will be clear AI really has outdone human mathematicians.

    *Depending on your philosophy of mathematics.
    **A counterexample to the Riemann hypothesis, which no mathematician expects, might or might not show AI superiority, depending on what finding it involves.

  47. John Morales says

    KG:

    But talking to a friend who is a professor of pure maths, he says the important advances are generally about discovering/inventing* new kinds of mathematical objects or relations, and so far, AI has not shown any ability to do that.

    (The goalposts keep moving, no?)

    Presumably, this does not count: https://www.quantamagazine.org/updates/transformation/ though it is a indeed a new mathematical object.

    Elusive complex structure found for S6, the six-dimensional sphere
    September 10, 2026
    &nbsp
    A large language model has found an object that has eluded mathematicians since 1947, despite intense efforts to find it or prove it doesn’t exist.

    —

    Of course, this is a nascent field, ever-improving.
    cf. https://ramanujanmachine.com/publications/

  48. KG says

    John Morales@61,

    I said new kinds of mathematical object, which the shape found clearly is not. Come back to me about “moving goalposts” when an AI does indeed invent a new kind of mathematical object which produces an important advance – or when one proves the Riemann conjecture or determines whether P=NP*. I’ll wait. My mathematician friend says there is indeed widespread concern among his graduate students and colleagues – but thinks much of that is because a lot of them were relying on “the boring bits”, which AI is good at, for their future careers.

    Incidentally, another limitation of the Navier-Stokes proof has come to light, although it doesn’t mean the proof fails to meet the “Millennium Problems” criteria for Navier-Stokes: the example found requires an external force with properties unlikely to be found in nature, and it’s been shown that the same approach cannot be generalised to deal with the cases physicists are actually interested in. There’s something of a pattern, noticeable also in the “AIs broke out of a sandbox and autonomously hacked other companies” example: a fanfare claiming a great breakthrough, or siren warning of an existential threat – then as details emerge, the whole thing looks rather less revolutionary.

    *I don’t know about the remaining 3 Millennium Problems – Riemann and P=?NP are the ones for which I have some idea what the question is.

  49. John Morales says

    I said new kinds of mathematical object, which the shape found clearly is not.

    Gotta keep those goalposts are always receding, KG.

    A bit like saying AIs can’t do new types of chess openings.

    —

    […] or when one proves the Riemann conjecture or determines whether P=NP*.

    But not if one proves that either or both are nondeterminable?

    (A bit like https://en.wikipedia.org/wiki/Continuum_hypothesis in ZFC)

Leave a Reply