I'd like to believe Grant but I find this assault on mathematics to be dehumanizing and demoralizing. AI has destroyed a major source of "life's meaning" for many people who pursued mathematics. Imagine waking up one morning to find that an alien technology has solved the problem you had dedicated your life to (and many more to boot.) Not another mathematician but an alien technology. As a scientist, I don't want to end up as a laboratory assistant to an AI. I can take not being the smartest human in the room, but I'm pretty sure I don't want to spend my life serving a silicon master.
> Imagine waking up one morning to find that an alien technology has solved the problem you had dedicated your life to
I'm sympathetic to mathematicians but this does not paint them in a good light.
If your primary purpose is SOLVING the problem, you should be ecstatic! If your primary purpose was rent-seeking so you could play with puzzles*, you'd be very upset.
* I find this characterization overly harsh, but if you're mad about your problem being solved, maybe you weren't in it for the right reasons
I suggest listening to "Math Vs Machine" [1], which has a much more nuanced take. It clearly lays out two types of use of AI in math, one that helps humans have better understanding, and one just bulldozes problems without helping us get any insights. It also touches on the topic of "does it matter if the machine does things in ways humans cannot understand".
There is a difference between the process of solving a problem and a problem being solved. If I solve a problem myself, I understand it, I find out something new, I get to try lots of things. It's engaging and fun. If I tell you to go solve a problem, I can learn it afterwards but I miss out on the entire process.
Personally, I enjoy the process of solving problems. I do not enjoy wrangling unpredictable AI agents.
Mathematics is more akin to art or a sport to many — philosophy, too. I don’t see that as a bad thing.
But there’s always a risk that someone in an adjacent domain makes a breakthrough that completely reshapes the research landscape overnight. It’s happened before AI and it will happen at some point again in years to come. Grief is understandable, but this isn’t the first time or the last time.
I wonder how many math professors think they were actively working on well-known unsolved problems. Is it just a small elite club, or is it every tenured / tenure-track professor in all R1 schools?
I'm not entirely sure this is fair. My belief is that mathematics exists in the intersection between art and science. I don't believe in progress for its own sake, so I'm inclined to say those interested should undertake it for the art, or to "play with puzzles." The rent-seeking argument and "weren't in it for the right reasons" is the same argument as calling musicians sellouts when they start making money. Making money doesn't make you a sellout, but it can.
A lot of people went into maths because they liked not only the results but the process of research and discovery. That has been taken from them. No matter how you feel about AI, I think we should all pause and at least think about this and validate those feelings.
Look at the most upvoted comment on that thread:
-----------
I feel like the point is still being missed, and it's very similar to the way software engineers are trying to cope. The argument you are making "Well yes now this part of math we will not need to do again, but now we can do this instead" however the issue with most people just don't want to do that....
Like who are we kidding, the type of nerds who were attracted to math and software were into it because we saw it as an art-form, it was a craft, it was being the only one in your city to be able to solve a complicated math problem, it was immortalizing yourself because you discovered a proof and added to the ever expanding human knowledge.
Imagine a leather workman who's life passion is making insanely high quality handbags, who takes joy in knowing everything he touches will last someone a lifetime and give them joy everytime they use it. And then not gradually, but almost overnight someone comes to his door and tells him we can make bags 50 times better than you at 1/tenth the price. But don't you worry, you can make more money overseeing the factory, or selling people on how good this bag is, and explaining that to them.
I mean, ok, that just is not what I wanted to do with my life. I never wanted to be a manager, I never wanted to be a salesman, I never wanted to be a writer. A craft I enjoy doing has been taken away from me, I should be allowed to mourn, I don't care that I can do something else since I never liked doing it in the first place. I enjoyed solving problems and writing code, I don't care if people say "well thats never what programming was about it was actually about becoming a manager all along" I never wanted to be a manager, I never wanted to manage, I never wanted to be a proof reader.
It's completely natural to be attached to the process of your job, rather than the outcome. It's not a phenomenon restricted to mathematics. We can say it shouldn't be like that, but it is.
> Imagine waking up one morning to find that an alien technology has solved the problem you had dedicated your life to
But it could be another human that did that, and the impact would be no different.
As a software engineer this is part of my life, and always has been. I may have been working on a software product for the last 3 years, and suddenly one morning I find out that another company has released a similar product, and selling it cheaper, now the value of my work is slashed by 90%. I may even have to drop my project and find something else to work on. This is a normal part of life.
The egos of these mathematicians is not what is important here, and they should get over themselves. The goals and values should be about advancing mathematics and, by proxy, humanity. It’s not about them.
That said, change is difficult for most. Let’s give them grace and have patience as they work through the stages of grief. Once they come to accept this new reality, let’s hope they can learn to embrace this new technology to continue the work of advancing the frontier.
I'm sympathetic to the mathematicians and don't like what AI is doing. Every other profession that relies on knowledge or thinking should be watching the mathematics community as a portent of the storm to come. OpenAI has extensively thought about and planned their disruption and chose mathematics to start with for a reason - small community, verifiable to some extent, can be used to improve model training, not seen positively by many average people etc. The AI companies are going to come for every field one by one.
I as a non maths person am not sure what you mean. Isn't AI solving lots of problems advancing the field? I understand that that maybe demoralising, but otherwise what do you mean by "assault on mathmematics"?
Solving problems and advancing the field are two related but separate endeavors.
Nobody has solved the Riemann Hypothesis, but it has been highly fertile ground for new mathematics. Answering the question is of far less value than the question itself.
1. One view of the field treats Maths as directly solving real-world problems. Under that lens, yes, solving problems is directly advantageous for the field. Even so, it's not necessarily a good thing for the participants. Suppose the incentive structures require tackling hard, long problems, the field attracts people who enjoy that sort of thing, it's hard a-priori to tell a hard problem from an aggrandized brute-force proof search, etc. Even if directly producing proofs is the desired output for the field, a capricious machine occasionally solving 40% of the big problems the field is working on is still demoralizing. On a human level, if a society has organized around producing 100yr+ problems and bestowed on you the privilege of tackling a few of them having the last 40yrs of your work solved by a fancy calculator is never going to feel good. Sure, that means that in retrospect that was a problem we could have waited to devote human effort toward, but you couldn't possibly have known that at the time. You worked hard, made locally correct decisions, and later found out that the work was worthless. Beyond that, I haven't seen any model of "Maths 2.0" produced where that issue goes away. We still want people to devote their lives to hard problems. The machine will still advance and solve a moderate fraction of them at random, killing many of the field's life's work. What motivation do you have in that environment to pursue those challenges in the first place? Why not pursue adtech or fintech and make the world a worse place instead?
2. "Solving problems" isn't directly the point of most of the field. That's true both from the perspective of most problems requiring many intermediate results (solving one useful thing requires solving many less useful things) and from the perspective of building the machinery capable of solving problems being much more important than the solutions themselves (a properly presented proof leveraging clever, new techniques is infinitely more valuable than whichever thing it proved -- indeed, the Millennium Prize problems were chosen precisely because it was assumed they would create such techniques, and for the human-solved ones they did -- without new techniques, eventually the field caps out at whichever frontier of knowledge we're currently stuck at).
3. A third view of the field treats it more like art. A machine producing behind your back the painting you were envisioning, or at least a close enough simile to fool the masses, doesn't advance "art" in any meaningful sense. It's extremely demoralizing when flaky cast tools are passed off for proper forged variants (your work is better, and it's sad to see people choosing a wrench they'll later snap in half instead), and when at a creative level people are accepting mediocrity instead of something objectively better.
So there's an interesting question about what mathematics is that the AI solution space is driving a wedge into and cracking wide, where previously the cracks were visible but could be ignored.
When is a problem in mathematics "solved?"
There's an old joke in math and computer science about a huge leap of logic in a proof either being flagged as "left as an exercise for the reader" or "trivially shown." The problem being that things are only trivial if you have all the context to know why they're trivial.
We can claim a problem is solved if there is one paper explaining the solution with airtight logic that is understood by five people on the planet. And that is the kind of solution that LLMs can kick out (by basically, to hand wave a lot of details that I will admit to not fully understanding, doing a search across the whole problem domain with vastly more capacity than most human beings to do the symbolic manipulation that makes up the mechanics of mathematics). But what does that really mean? To have something "proven," but only to the comprehension of five people on the planet?
3blue1brown is making the interesting argument that perhaps mathematics practice shifts from a frontier that AI might actually be better at exploring than humans, to the process of digesting the output of that exploration and making it sensible and comprehensible to the billions of mines that could take advantage of that knowledge... If it were shaped in a way that could stick in their brains.
That's a kind of creativity that AI is not likely to be good at in the near future, because AI doesn't really "understand people."
How is that any different than another kind of automation?
AI will make mathematics so much more approachable, comprehensible, and applicable as more problems gets solved. If the costs is a handful of PhDs tenure track, so be it. Academics didn’t much to seem to mind when that happens to blue collar workers.
Nothing stopping anyone from studying mathematics here.
I'm not sure that's a fair summary of the situation though. If the machines understand math but humans don't, we're going to be in a bad bad bad spot.
We need humans to help guide all this, need to learn all we can.
I agree with Sanderson that the opportunity seems larger than in the past. The challenges are quite different, but if somebody loves math, loves learning about it, like I do, I see so much opportunity.
There's going to be so many more connections to evaluate, so much work needed for unification and simplification and pedagogy, and above all there's going to be intense demand for mathematical taste.
Well that's the point, there's immense value for human involvement in these skills. More now than in the past. If you can do more with math, additional funding avenues open up.
Jane Street will be using tons of AI in math, but boy are they going to need a lot of humans to try to keep it all under control.
I find myself sometimes wondering if the folks on Star Trek would consider themselves laboratory assistants to an AI, seeing as how the ship computer kind of works like an LLM: interprets voice queries into actions and takes actions, queries the database for answers, can extrapolate when asked to depending on how the writers are feeling... That kind of thing.
> AI has destroyed a major source of "life's meaning" for many people who pursued mathematics.
I studied mathematics and don't agree with this at all, for several reasons. For one the entire point of maths is that it describes a formal system in which one can 'shut up and calculate'. Mechanizing maths goes back to Leibniz and Hilbert. Nobody should surprised that the big formalism with explicit rules you invented is the first thing to yield to automation, that's why it exists.
Secondly research, including maths for a long time has been a collaboration of people, institutions, machines and processes, even maths does not consist of Faustian heroes in caves having revelations.
And thirdly just because a machine produces a 500 page proof doesn't mean your job is done. Science produces knowledge. Knowledge is information made legible to understanding by people. So there will be endless work for humans to go over machine proofs and turn it into something we understand. Just like chess computers ultimately expanded human chess.
> So there will be endless work for humans to go over machine proofs and turn it into something we understand
I'm not convinced humans will be better at this than LLMs. Frontier models are quite good at handling even deliberately lazy/ambiguous eli5 prompts on complex topics IME.
AI solve my problem where I have to hand code all the time. I'm happier than ever. Math's analogy is they now get to find problems instead of solving them, then get LLM's to solve them or solve them together. I've been saying this, numbers go up, and LLM makes the amount of problems to solve go up, not less.
I think a lot of "mathematicians" are saddened by the sudden realization they are not even interested in it and were only in it for status and prestige. A lot of them are such pretentious tools that they willingly wasted their lives doing the absurd PhD rituals. Academia is dead.
What about the people who are in it to think deeply about mathematics questions? It's a lot harder to feel like that's worth doing when there's a black box with a crank that takes money in and spits out mathematics answers.
Like calling Lee Sedol a pretentious tool for retiring in the face of AlphaGo. If you really do mathematics for the enjoyment of doing mathematics, following a roughly unbroken line of tradition for a couple hundred years, an alien computer mind showing up and smashing everything would perhaps be upsetting, don't you think?
I have been feeling this way about software. Don't get me wrong, there are some things I am very interested in and would work on for recreation (macros!), but a lot of it I'm less interested in.
Frankly, despite the obvious financial worries, it's been a little freeing. I can go do something else (because, uh, I may have to).
I actually remember that in an old interview (that I can't find sorry) Grant Sanderson said he moved in to YouTube because he thought it was a more promising career than being mathematician, it's interesting to see his opinion change.
Hmm, that seems like a pretty cynical point to make. He still deeply is in the field of mathematics, as a math educator. He even makes this exact point in this video, that communicating math is still a hard and respectable part of the profession.
As a math educator doing what you love you requires an audience.
Its kind of in Grant's incentives to push the interest for mathematics, because without an audience he won't be able to keep doing what he loves.
Like, my old uni has a history department, and had an Egyptian history teacher. This was at some point deemed a pointless use of resources because there was no role in society except for teacher of the same subject after that, and the classes shrunk, and she ended up getting offered a role in admin. She was not happy in that role.
I remember a post in... Reddit I think it was? where I commented on an early video discussion that he should teach online full time, because that kind of intuitive education was sorely needed. He actually answered the comment, along the lines of "that would be cool but I already have my hands full in $UNIVERSITY".
I think it was relatively soon after that interaction that he started doing youtube full time. I've always wanted to think I made a teeny tiny contribution to move the needle there :)
Yeah. All due respect to Grant Sanderson, he is both great scientist and great entrepreneur, but getting an entry-career advice from a multi-millionaire is probably not quite the best idea out of all ideas in a popular discourse. Especially if the field in question was famous for producing posts discouraging people from going all-in in it, for many many years even before AI craze.
First, "cognitive labor" is a form of labor - it's different vs. creative work, even though both use your mind. If your math theorem can be written up in partnership with an LLM, wow! You can be done with that, and go do other great work. Cognitive labor is a form of labor. Let's talk about physical labor and leverage to understand how removing labor translates to time. On your farm you buy a tractor to replace your backhoe. If it is a hobby farm you get in the tractor and you're done super fast. Or, you can take that tractor, buy a bigger farm, and spend the same amount of time you'd have spent backhoeing, but in a far more leveraged way. You can trade removed labor for time in a way that suits you.
Second; the joyful moments we get from finding generalizable abstractions and nice solutions are an adaptation. Humans can climb infinitely high conceptual hierarchies - we'll never stop climbing, using the previous ones as tools to find the next. As we do, we gain joy with each discovery because. We get so much done with brains that run on 20 watts - are we tuned to receive a neurochemical reward for finding cognitive efficiency? If AI means more concepts are discovered the tower gets taller - and we find those joyful moments more often. If there are infinite moments to be found, why not go as fast as possible!
Third; everyone else is doing it, it's a baseline skill. Knowing how to work without AI is like knowing how to ride a horse. Useful and fun! Bedsides - only 150ish years ago horses were the only viable transport. Cars were new; toys for the rich. But cars got so good so fast that we eventually reconfigured whole cities. Your homework is to consider why and how horses got disrupted.
New divisions are emerging in the lump of roles humans had at work. AI takes away roles from your job and lets us do more with less. That's identical to every mechanization before it. It's killing some roles that make up your work, even as it creates new ones.
There's so much to be excited about. Fear is a rational response to change. But so is hope :)
> If your math theorem can be written up in partnership with an LLM, wow! You can be done with that, and go do other great work.
If your painting can be painted in partnership with an LLM, wow!
If your essay can be written in partnership with an LLM, wow!
If your painting portfolio website can be written in partnership with an LLM, wow!
If your museum show can be designed in partnership with an LLM, wow!
If the placards next to each of your paintings can be written in partnership with an LLM, wow!
If the car that takes me to the gallery can be designed and driven in partnership with an LLM, wow!
If the newspaper or novel that you read on the way to the gallery can be written in partnership with an LLM, wow!
If an LLM can give tours at the gallery and answer questions asked by visitors, wow!
If an LLM can write the music that I listen to on my way out of the gallery, wow!
If an LLM can synthetically perform the music, too, wow!
> If AI means more concepts are discovered the tower gets taller - and we find those joyful moments more often. If there are infinite moments to be found, why not go as fast as possible!
I don't understand how a person can say this with a straight face. There has been so much discussion of this, and so many reasons have been given, that one can only assume that, if you're asking the question and you have thought about the issue and read what people have said, you're not only asking rhetorically but have some personal reason for insisting on a frankly pretty frustrating, borderline insensitive insistence on being positive. I'm not trying to be snotty or rude. I just want to say that to me there is something uncanny about your comment, in its total blindness to any of the reasons someone might not only disagree with you but grieve the death of, say, their academic discipline in its current form -- and in its weirdly breezy lack of empathy.
Are you suggesting that if an AI could do all of those things to your satisfaction you would not be extremely impressed.
I do have empathy for people who see the world changing and are forced to adjust. That can be uncomfortable, but it is something that is inevitable with the changing world. It has always been this way, There is very little need to make swords these days, and as anything becomes superseded there are people that are attached to the old things and there are people dependent on the old things.
I think it entirely possible to have sympathy for their position while disagreeing that the solution is to prevent change.
I don't see this as a breezy lack of empathy. I see this as a hand reaching out and saying 'Hey, you can come this way and things aren't so bad after all'
> AI takes away work and lets us do more with less - just like every mechanization that came before!
No, it's not the same as mechanization. The fundamental difference, and why I find it illogical to argue either for "AI will open up tons of new work avenues" or "AI will kill all jobs" is that it all hinges on whether AI "caps out" at an intelligence level below humans (in at least some domains/capabilities), or if it is able to surpass human intelligence at all levels.
If the former is true, then yes, there will be a lot for humans to do. If the latter is true, then no, there will be no reasonable cognitive role for humans to play (beyond perhaps for some legal, "nostalgic" reasons or what not). So whether one decides to be optimistic or pessimistic really depends on if you think AI will stay at a level below humans. I really don't know, so I'm not "taking sides" in this debate.
On one hand it is hard to fathom not being the most intelligent thing around.
On the other hand, if we are going to survive as a species for a long time it probably depends on us learning how to not be the most intelligent thing around.
Even if it leads to our demise, I don’t think willfully holding ourselves back is the way forward
> ...hinges on whether AI "caps out" at an intelligence level below humans (in at least some domains/capabilities), or if it is able to surpass human intelligence at all levels.
I agree with you, just wondering your thought on how we will know the latter has come true? As I understand it, you're describing ASI, not AGI. The way I personally define AGI, we are already there.
I think one form of cognitive labor will remain: What do we actually want to do/build. And this is simply because AI does not know our internal world. Maybe in a few years with some kind of BCI that is extremely good. But I don't see it happening at least for a few years.
> A good precedent is mechanized agriculture. When machines effectively replaced most humans in agriculture, unemployment did not rise to 90%.
I don't consider this a good precedent to the AI situation at all.
When this happened, humans went off and started doing what we now consider "knowledge work". They escaped massive unemployment by being smarter than the tractors. How do they escape AI job loss when AI becomes smarter than the average human? or smarter than the smartest human?
In prior displacements humans moved up the stack. There is no longer room for them at higher levels of the stack if AI beats them at cognitive labor.
And this is why we must ensure that the AIs get aligned towards humanity and crucially that they don't get aligned towards capital, which seems to be the current trajectory.
There's a better chance that a humanity-aligned AI would make sure that humans have something to do and keep us alive versus a capital-aligned one. At the same time, if AI would be the deathblow to capitalism, that'd be very, _very_ ironic.
Reflecting on that a year later is interesting. Agents have become REALLY good since then, and I consider the "writing code" side of my profession to be largely annihilated. Some days that's really hard to accept.
But I think Grant's conclusion about math still largely echoes what I wrote about software: don't underestimate the social aspect of these activities, and if anything AI has made the social side more important. I deliberately avoided declaring, "now is the best time to learn to write code". I suspected that skill might be going out the window. But learning systems thinking and the social skills needed to connect people's dreams to reality continues to be an important art, an art I still have the privilege to practice every day.
And if we do reach the "anything you can do I can do better" stage, then, as Grant said, that's a hard to imagine world, and one where "it can write really good code" might be the least of our problems.
At this point I'm ready to call it... LLMs aren't going to ruin my job. They're amazing, they're useful... and the problem remains that to a first approximation human beings are uninterested in spending all day talking to a computer, whether through a classic IDE or through an LLM. Specifying a problem enough for an LLM to solve it, even if we stipulate one that does it perfectly, is already more than most people want to do.
You also aren't going to get an accountant running 4 simultaneous streams of AI-driven development and working through an AWS issue when something goes wrong.
The problem is you're focused on 1st order effects. I presume you work at some company that sells a product or provides a service - what happens if the demand for that evaporates because the clients can do it themselves or the market has been flooded by much cheaper alternatives created using AI? Or people who bought the product can't do it as they lost their jobs to automation?
Second, if "all software vendors in all spaces get wiped out" was going to happen I think we'd be seeing more of it already. Yes, there is support for some software vendors possibly getting wiped out... but, then again, you can't look at a bad vibe coded implementation of "the entire Adobe suite" that was rather overblown in the press coverage and forget that Adobe also has AIs. Instead what I'm seeing at my own workplace as we ponder the question of "should we replace this service we buy with some local vibe coding" is still remaining but we don't want to support that. Can we replace Confluence with vibe coding? Yes, but that was true five years ago too and we didn't want to do it then, either. We still don't want to do it. HN commenters have always overestimated the desire everyone in the world has to become programmers and take responsibility for everything.
One wonders what percentage of those who are certain that vibe coding is going to nuke all software vendors can't even be arsed to self-host a Jellyfin or VaultWarden install... which is way easier than vibe coding the whole of those things from scratch, and we're talking about people who are already programmers. Physician, heal thyself.
I agree. I think the days of the developer who hated talking to anyone but is still very valuable are over.
I think social skills are much more important today because one person can do so much more but this person needs to talk to many others to know what to build.
And this is why I am likely to leave software. I actually really like people, but I am autistic and not especially good at talking to them, and it's quite drianing to do so.
I will miss coding, but I had a good run. Time to learn a trade. I don't begrudge this fate.
No… the person who's going to do well is the one who doesn't have to ask what to build. The one who has to ask is no better than an AI, except they start out wanting to ask, rather than starting out with no interest whatsoever until they're prompted to have one.
When everything you ask for is instant and valueless, even asking becomes meaningless, and everyone is waiting to see something they'd not thought of.
Then, a thousand thousand copies, immediately… but also, watching that person like a hawk to see if they can do it again. That's the future. A lot of people will just plain become unable to think or intend anything on their own, or even want anything.
I think a lot of these reactions by mathematicians are more akin to group theory than anything else. Reality is that things will naturally sort themselves out and there will be some role mathematicians play in all of it (most likely). To me it seems the timeline that some skill essentially gets retired such that if you want to keep doing it is more like "historical re-enactment" is getting quite short. Things like manually coding something is going to be a hobby. Some aspects of math are going to be a hobby, and professional math will be all about pushing up against the edge of what we can do mathematically with AI.
3blue1brown has the absolute most educational and entertaining videos. Always fascinated by his animations which are custom programmed by him. I also appreciate the diversity of topics covered from deep math to software to science.
This is the same naivete that I saw in coding some time ago before most people finally accepted there is no going back and the times of writing code by hand are gone.
And the guy is not a mathematician primarily - he's a Youtube creator and entrepreneur. I liked a lot of his videos but now I usually just ask the LLM for detailed explanation and I can probe it as much as I want if something is not clear enough.
Education will also take a huge hit, you can't really beat an almost free supergenius that has all the time in the world to explain something to you in as much detail as you wish.
It is not, and for the same reason it's not over for mathematics. For anyone doing anything truly novel, and yes there are still software engineers out there doing that, it's still inept.
It's de jour to leverage formal verification as LLMs are quite good at looping it, but that's still a tool we had to invent to solve problems. It'll be truly over when we're being outpaced in the realm of invention but I have seen absolutely nothing to suggest that has meaningfully occurred at all. Coding by hand is not gone, there's just less of a reason to do so now.
I’m a cheerleader for the Khanmigo AI tutor effort. But, apparently so far they are discovering that the hard part is getting the kids motivated to actually ask the AI any questions.
3Blue1Brown does an amazing job of making math interesting to learn in itself. As opposed to the hundreds of millions of kids told “You just have to learn it or you’ll be dumb. Keep up or you’ll fall behind forever. Go!”
The vid calls out explicitly that math education is a primarily a motivation problem and I agree. One way or another, we need a lot more focus on that.
I can beat this “supergenius” - I hire a professor that knows exactly how to teach and what to teach. This professor isn’t rolling the dice every time I ask them a question. Boom.
There's never been a better time to be a professional mathematician, perhaps. Just like there's never been a better time to be a senior software engineer.
An 18 year old entering university today will retire somewhere in the 2070s. Is mathematics, or software engineering, a promising career path for the next 50 years? No, it's most certainly not, and anyone suggesting otherwise is delusional. I don't know what would be a good career path, but most certainly avoid anything that AI is having the most success at in 2026.
Too early to tell, but IMO unlikely. Let's see what his take will be once an AI can propose a topic, write the script, animate, and narrate a good educational math video.
He won't care because he has established position and is set. It will be just like everybody else - if the individual is at the top of the food chain then he/she is all over it how you are using it wrong and how you should use it (don't forget to follow and subscribe).
Yep, that's what I see in the CS field -- people at the top are advocating AI and don't understand why the hell the others are hesitant. I meant the top engineers, not the management. I look up to them because of their skills, but that doesn't mean I need to agree with everything they said.
But you already have Claude-directed videos that utilize OP's manim project (which he uses himself to do 3b1b) that anyone else can create. Grant open-sourced manim and now anyone can point a model at it to create their own 3b1b-like videos. They're popping up all over the place as Opus 5.5 demo videos just in the last few weeks. So... AGI?
I tried watching 3b1b over the years, really wanted to enjoy it. But I never found the way he presents things intuitive and never got a palpable intuition that helped me with math classes or work.
I saw some clearly AI slop made videos and got much better understanding of topics, since they don't do foreshadowing for an awe-inspiring connection that we'd admire together after 15 min videos, the clankers just get to the point. Not claiming those videos are too useful either, but I did however see accelerated learning gains from asking LLMs to explain (and do Bret Victor-style small interactive demos) math and algos, something I have never experienced from any of the educational STEM Youtube channels.
Even if you find them intuitive, they produce no learning gains for anyone. Take a lin alg test, watch the 'Essence of lin alg' videos, take comparable test again. I guarantee you won't see any change even if you 'understood matrices better'. They are pure mathporn.
This is not my experience. Having an intuitive grasp of e.g. the graphical interpretation of the determinant instantly lets you figure out most of the key determinant properties without having to memorize them. That's the primary thing you get from intuition imo, having a sense of what must be true without having to memorize or prove it.
No learning gains? I teach linear algebra. I can assure you that students who watch and study the "Essence of linear algebra" videos (which I recommend on the syllabus) do better on my tests that those who don't, because they learn stuff from them.
I'm actually quite optimistic, was pointing out that, at least to to me, AI slop already produces better results than big educational content creators. For something concrete, even if not classic math, I credit direct LLM interaction highly for accelerating my understanding of CPU design and compiler tech.
fwiw the fact that brains work fundamentally differently was my big penny dropping realization after digging into what LLMs are doing. Everything (I assert) a brain does (the function between inputs and outputs) is due to training data. There is no "instinct" and there is no "ROM-driven way to think" in any brain. It's training data all the way down. As a consequence different brains do indeed think in fundamentally different ways.
I suspect people who are deciding what to study will do so on the basis of a multifactor view of what interests them, how they predict their career will be after finishing said study etc.
There ought to be people watching his channel who wonder whether given all these AI advances math is even worth studying now.
In other news, there's never been a better time to be a software developer. Come to think of it, there's also never been a better time to invest in learning copy-editing and proof-reading as a side hustle. It's a Brave New World. Everything is fine, the demand just got expanded, the pie got bigger, nothing more. New doors will be opened, nothing has changed; in fact, everything improved. Stochastic parrots are no match for our carbon-based g.
I doubt demand expanded much. I mean there is just that many games/whatever apps that the whole humanity needs. Maybe ti will go up a bit, like 50%, but getting 10x (or very easily 1000x if we just vibe code them) in supply doesn't change that.
I mean if you read some history it’s not the first time complete workforces have been replaced. It doesn’t mean the same jobs stay, but usually there will be something to do.
I'm sympathetic to mathematicians but this does not paint them in a good light.
If your primary purpose is SOLVING the problem, you should be ecstatic! If your primary purpose was rent-seeking so you could play with puzzles*, you'd be very upset.
* I find this characterization overly harsh, but if you're mad about your problem being solved, maybe you weren't in it for the right reasons
[1] https://radiolab.org/podcast/math-vs-machine
Personally, I enjoy the process of solving problems. I do not enjoy wrangling unpredictable AI agents.
But there’s always a risk that someone in an adjacent domain makes a breakthrough that completely reshapes the research landscape overnight. It’s happened before AI and it will happen at some point again in years to come. Grief is understandable, but this isn’t the first time or the last time.
There’s something here about the journey not just the destination.
Press a button and problem solved is not too fun
Look at the most upvoted comment on that thread:
-----------
I feel like the point is still being missed, and it's very similar to the way software engineers are trying to cope. The argument you are making "Well yes now this part of math we will not need to do again, but now we can do this instead" however the issue with most people just don't want to do that....
Like who are we kidding, the type of nerds who were attracted to math and software were into it because we saw it as an art-form, it was a craft, it was being the only one in your city to be able to solve a complicated math problem, it was immortalizing yourself because you discovered a proof and added to the ever expanding human knowledge.
Imagine a leather workman who's life passion is making insanely high quality handbags, who takes joy in knowing everything he touches will last someone a lifetime and give them joy everytime they use it. And then not gradually, but almost overnight someone comes to his door and tells him we can make bags 50 times better than you at 1/tenth the price. But don't you worry, you can make more money overseeing the factory, or selling people on how good this bag is, and explaining that to them.
I mean, ok, that just is not what I wanted to do with my life. I never wanted to be a manager, I never wanted to be a salesman, I never wanted to be a writer. A craft I enjoy doing has been taken away from me, I should be allowed to mourn, I don't care that I can do something else since I never liked doing it in the first place. I enjoyed solving problems and writing code, I don't care if people say "well thats never what programming was about it was actually about becoming a manager all along" I never wanted to be a manager, I never wanted to manage, I never wanted to be a proof reader.
I just hate this existence.
But it could be another human that did that, and the impact would be no different.
As a software engineer this is part of my life, and always has been. I may have been working on a software product for the last 3 years, and suddenly one morning I find out that another company has released a similar product, and selling it cheaper, now the value of my work is slashed by 90%. I may even have to drop my project and find something else to work on. This is a normal part of life.
That said, change is difficult for most. Let’s give them grace and have patience as they work through the stages of grief. Once they come to accept this new reality, let’s hope they can learn to embrace this new technology to continue the work of advancing the frontier.
Nobody has solved the Riemann Hypothesis, but it has been highly fertile ground for new mathematics. Answering the question is of far less value than the question itself.
1. One view of the field treats Maths as directly solving real-world problems. Under that lens, yes, solving problems is directly advantageous for the field. Even so, it's not necessarily a good thing for the participants. Suppose the incentive structures require tackling hard, long problems, the field attracts people who enjoy that sort of thing, it's hard a-priori to tell a hard problem from an aggrandized brute-force proof search, etc. Even if directly producing proofs is the desired output for the field, a capricious machine occasionally solving 40% of the big problems the field is working on is still demoralizing. On a human level, if a society has organized around producing 100yr+ problems and bestowed on you the privilege of tackling a few of them having the last 40yrs of your work solved by a fancy calculator is never going to feel good. Sure, that means that in retrospect that was a problem we could have waited to devote human effort toward, but you couldn't possibly have known that at the time. You worked hard, made locally correct decisions, and later found out that the work was worthless. Beyond that, I haven't seen any model of "Maths 2.0" produced where that issue goes away. We still want people to devote their lives to hard problems. The machine will still advance and solve a moderate fraction of them at random, killing many of the field's life's work. What motivation do you have in that environment to pursue those challenges in the first place? Why not pursue adtech or fintech and make the world a worse place instead?
2. "Solving problems" isn't directly the point of most of the field. That's true both from the perspective of most problems requiring many intermediate results (solving one useful thing requires solving many less useful things) and from the perspective of building the machinery capable of solving problems being much more important than the solutions themselves (a properly presented proof leveraging clever, new techniques is infinitely more valuable than whichever thing it proved -- indeed, the Millennium Prize problems were chosen precisely because it was assumed they would create such techniques, and for the human-solved ones they did -- without new techniques, eventually the field caps out at whichever frontier of knowledge we're currently stuck at).
3. A third view of the field treats it more like art. A machine producing behind your back the painting you were envisioning, or at least a close enough simile to fool the masses, doesn't advance "art" in any meaningful sense. It's extremely demoralizing when flaky cast tools are passed off for proper forged variants (your work is better, and it's sad to see people choosing a wrench they'll later snap in half instead), and when at a creative level people are accepting mediocrity instead of something objectively better.
When is a problem in mathematics "solved?"
There's an old joke in math and computer science about a huge leap of logic in a proof either being flagged as "left as an exercise for the reader" or "trivially shown." The problem being that things are only trivial if you have all the context to know why they're trivial.
We can claim a problem is solved if there is one paper explaining the solution with airtight logic that is understood by five people on the planet. And that is the kind of solution that LLMs can kick out (by basically, to hand wave a lot of details that I will admit to not fully understanding, doing a search across the whole problem domain with vastly more capacity than most human beings to do the symbolic manipulation that makes up the mechanics of mathematics). But what does that really mean? To have something "proven," but only to the comprehension of five people on the planet?
3blue1brown is making the interesting argument that perhaps mathematics practice shifts from a frontier that AI might actually be better at exploring than humans, to the process of digesting the output of that exploration and making it sensible and comprehensible to the billions of mines that could take advantage of that knowledge... If it were shaped in a way that could stick in their brains.
That's a kind of creativity that AI is not likely to be good at in the near future, because AI doesn't really "understand people."
AI will make mathematics so much more approachable, comprehensible, and applicable as more problems gets solved. If the costs is a handful of PhDs tenure track, so be it. Academics didn’t much to seem to mind when that happens to blue collar workers.
Nothing stopping anyone from studying mathematics here.
We need humans to help guide all this, need to learn all we can.
I agree with Sanderson that the opportunity seems larger than in the past. The challenges are quite different, but if somebody loves math, loves learning about it, like I do, I see so much opportunity.
There's going to be so many more connections to evaluate, so much work needed for unification and simplification and pedagogy, and above all there's going to be intense demand for mathematical taste.
With potentially plummeting value for human involvement in these skills, I think that’s precisely what many people are worried about.
Jane Street will be using tons of AI in math, but boy are they going to need a lot of humans to try to keep it all under control.
To generalize beyond math... how many bits of working memory do you have? How about a transformer?
— Ojibwe proverb
I studied mathematics and don't agree with this at all, for several reasons. For one the entire point of maths is that it describes a formal system in which one can 'shut up and calculate'. Mechanizing maths goes back to Leibniz and Hilbert. Nobody should surprised that the big formalism with explicit rules you invented is the first thing to yield to automation, that's why it exists.
Secondly research, including maths for a long time has been a collaboration of people, institutions, machines and processes, even maths does not consist of Faustian heroes in caves having revelations.
And thirdly just because a machine produces a 500 page proof doesn't mean your job is done. Science produces knowledge. Knowledge is information made legible to understanding by people. So there will be endless work for humans to go over machine proofs and turn it into something we understand. Just like chess computers ultimately expanded human chess.
I'm not convinced humans will be better at this than LLMs. Frontier models are quite good at handling even deliberately lazy/ambiguous eli5 prompts on complex topics IME.
Like calling Lee Sedol a pretentious tool for retiring in the face of AlphaGo. If you really do mathematics for the enjoyment of doing mathematics, following a roughly unbroken line of tradition for a couple hundred years, an alien computer mind showing up and smashing everything would perhaps be upsetting, don't you think?
Then I finally realized that Professors spend most of their days in meetings and management - not actual research!
Frankly, despite the obvious financial worries, it's been a little freeing. I can go do something else (because, uh, I may have to).
Its kind of in Grant's incentives to push the interest for mathematics, because without an audience he won't be able to keep doing what he loves.
Like, my old uni has a history department, and had an Egyptian history teacher. This was at some point deemed a pointless use of resources because there was no role in society except for teacher of the same subject after that, and the classes shrunk, and she ended up getting offered a role in admin. She was not happy in that role.
I think it was relatively soon after that interaction that he started doing youtube full time. I've always wanted to think I made a teeny tiny contribution to move the needle there :)
First, "cognitive labor" is a form of labor - it's different vs. creative work, even though both use your mind. If your math theorem can be written up in partnership with an LLM, wow! You can be done with that, and go do other great work. Cognitive labor is a form of labor. Let's talk about physical labor and leverage to understand how removing labor translates to time. On your farm you buy a tractor to replace your backhoe. If it is a hobby farm you get in the tractor and you're done super fast. Or, you can take that tractor, buy a bigger farm, and spend the same amount of time you'd have spent backhoeing, but in a far more leveraged way. You can trade removed labor for time in a way that suits you.
Second; the joyful moments we get from finding generalizable abstractions and nice solutions are an adaptation. Humans can climb infinitely high conceptual hierarchies - we'll never stop climbing, using the previous ones as tools to find the next. As we do, we gain joy with each discovery because. We get so much done with brains that run on 20 watts - are we tuned to receive a neurochemical reward for finding cognitive efficiency? If AI means more concepts are discovered the tower gets taller - and we find those joyful moments more often. If there are infinite moments to be found, why not go as fast as possible!
Third; everyone else is doing it, it's a baseline skill. Knowing how to work without AI is like knowing how to ride a horse. Useful and fun! Bedsides - only 150ish years ago horses were the only viable transport. Cars were new; toys for the rich. But cars got so good so fast that we eventually reconfigured whole cities. Your homework is to consider why and how horses got disrupted.
New divisions are emerging in the lump of roles humans had at work. AI takes away roles from your job and lets us do more with less. That's identical to every mechanization before it. It's killing some roles that make up your work, even as it creates new ones.
There's so much to be excited about. Fear is a rational response to change. But so is hope :)
If your painting can be painted in partnership with an LLM, wow!
If your essay can be written in partnership with an LLM, wow!
If your painting portfolio website can be written in partnership with an LLM, wow!
If your museum show can be designed in partnership with an LLM, wow!
If the placards next to each of your paintings can be written in partnership with an LLM, wow!
If the car that takes me to the gallery can be designed and driven in partnership with an LLM, wow!
If the newspaper or novel that you read on the way to the gallery can be written in partnership with an LLM, wow!
If an LLM can give tours at the gallery and answer questions asked by visitors, wow!
If an LLM can write the music that I listen to on my way out of the gallery, wow!
If an LLM can synthetically perform the music, too, wow!
> If AI means more concepts are discovered the tower gets taller - and we find those joyful moments more often. If there are infinite moments to be found, why not go as fast as possible!
I don't understand how a person can say this with a straight face. There has been so much discussion of this, and so many reasons have been given, that one can only assume that, if you're asking the question and you have thought about the issue and read what people have said, you're not only asking rhetorically but have some personal reason for insisting on a frankly pretty frustrating, borderline insensitive insistence on being positive. I'm not trying to be snotty or rude. I just want to say that to me there is something uncanny about your comment, in its total blindness to any of the reasons someone might not only disagree with you but grieve the death of, say, their academic discipline in its current form -- and in its weirdly breezy lack of empathy.
Are you suggesting that if an AI could do all of those things to your satisfaction you would not be extremely impressed.
I do have empathy for people who see the world changing and are forced to adjust. That can be uncomfortable, but it is something that is inevitable with the changing world. It has always been this way, There is very little need to make swords these days, and as anything becomes superseded there are people that are attached to the old things and there are people dependent on the old things.
I think it entirely possible to have sympathy for their position while disagreeing that the solution is to prevent change.
I don't see this as a breezy lack of empathy. I see this as a hand reaching out and saying 'Hey, you can come this way and things aren't so bad after all'
No, it's not the same as mechanization. The fundamental difference, and why I find it illogical to argue either for "AI will open up tons of new work avenues" or "AI will kill all jobs" is that it all hinges on whether AI "caps out" at an intelligence level below humans (in at least some domains/capabilities), or if it is able to surpass human intelligence at all levels.
If the former is true, then yes, there will be a lot for humans to do. If the latter is true, then no, there will be no reasonable cognitive role for humans to play (beyond perhaps for some legal, "nostalgic" reasons or what not). So whether one decides to be optimistic or pessimistic really depends on if you think AI will stay at a level below humans. I really don't know, so I'm not "taking sides" in this debate.
On the other hand, if we are going to survive as a species for a long time it probably depends on us learning how to not be the most intelligent thing around.
Even if it leads to our demise, I don’t think willfully holding ourselves back is the way forward
I agree with you, just wondering your thought on how we will know the latter has come true? As I understand it, you're describing ASI, not AGI. The way I personally define AGI, we are already there.
Sure, but you don't need very many people to do that relative to the billions of workers that currently exist.
If you wipe out 90% of all jobs instead of 100% the practical implications for society as a whole are roughly exactly the same.
The Great Depression was "only" 25% unemployment.
I don't consider this a good precedent to the AI situation at all.
When this happened, humans went off and started doing what we now consider "knowledge work". They escaped massive unemployment by being smarter than the tractors. How do they escape AI job loss when AI becomes smarter than the average human? or smarter than the smartest human?
In prior displacements humans moved up the stack. There is no longer room for them at higher levels of the stack if AI beats them at cognitive labor.
There's a better chance that a humanity-aligned AI would make sure that humans have something to do and keep us alive versus a capital-aligned one. At the same time, if AI would be the deathblow to capitalism, that'd be very, _very_ ironic.
Reflecting on that a year later is interesting. Agents have become REALLY good since then, and I consider the "writing code" side of my profession to be largely annihilated. Some days that's really hard to accept.
But I think Grant's conclusion about math still largely echoes what I wrote about software: don't underestimate the social aspect of these activities, and if anything AI has made the social side more important. I deliberately avoided declaring, "now is the best time to learn to write code". I suspected that skill might be going out the window. But learning systems thinking and the social skills needed to connect people's dreams to reality continues to be an important art, an art I still have the privilege to practice every day.
And if we do reach the "anything you can do I can do better" stage, then, as Grant said, that's a hard to imagine world, and one where "it can write really good code" might be the least of our problems.
You also aren't going to get an accountant running 4 simultaneous streams of AI-driven development and working through an AWS issue when something goes wrong.
No comment as to the next generation of AIs.
Second, if "all software vendors in all spaces get wiped out" was going to happen I think we'd be seeing more of it already. Yes, there is support for some software vendors possibly getting wiped out... but, then again, you can't look at a bad vibe coded implementation of "the entire Adobe suite" that was rather overblown in the press coverage and forget that Adobe also has AIs. Instead what I'm seeing at my own workplace as we ponder the question of "should we replace this service we buy with some local vibe coding" is still remaining but we don't want to support that. Can we replace Confluence with vibe coding? Yes, but that was true five years ago too and we didn't want to do it then, either. We still don't want to do it. HN commenters have always overestimated the desire everyone in the world has to become programmers and take responsibility for everything.
One wonders what percentage of those who are certain that vibe coding is going to nuke all software vendors can't even be arsed to self-host a Jellyfin or VaultWarden install... which is way easier than vibe coding the whole of those things from scratch, and we're talking about people who are already programmers. Physician, heal thyself.
I think social skills are much more important today because one person can do so much more but this person needs to talk to many others to know what to build.
I will miss coding, but I had a good run. Time to learn a trade. I don't begrudge this fate.
When everything you ask for is instant and valueless, even asking becomes meaningless, and everyone is waiting to see something they'd not thought of.
Then, a thousand thousand copies, immediately… but also, watching that person like a hawk to see if they can do it again. That's the future. A lot of people will just plain become unable to think or intend anything on their own, or even want anything.
And the guy is not a mathematician primarily - he's a Youtube creator and entrepreneur. I liked a lot of his videos but now I usually just ask the LLM for detailed explanation and I can probe it as much as I want if something is not clear enough.
Education will also take a huge hit, you can't really beat an almost free supergenius that has all the time in the world to explain something to you in as much detail as you wish.
It is not, and for the same reason it's not over for mathematics. For anyone doing anything truly novel, and yes there are still software engineers out there doing that, it's still inept.
It's de jour to leverage formal verification as LLMs are quite good at looping it, but that's still a tool we had to invent to solve problems. It'll be truly over when we're being outpaced in the realm of invention but I have seen absolutely nothing to suggest that has meaningfully occurred at all. Coding by hand is not gone, there's just less of a reason to do so now.
3Blue1Brown does an amazing job of making math interesting to learn in itself. As opposed to the hundreds of millions of kids told “You just have to learn it or you’ll be dumb. Keep up or you’ll fall behind forever. Go!”
The vid calls out explicitly that math education is a primarily a motivation problem and I agree. One way or another, we need a lot more focus on that.
There's never been a better time to be a professional mathematician, perhaps. Just like there's never been a better time to be a senior software engineer.
An 18 year old entering university today will retire somewhere in the 2070s. Is mathematics, or software engineering, a promising career path for the next 50 years? No, it's most certainly not, and anyone suggesting otherwise is delusional. I don't know what would be a good career path, but most certainly avoid anything that AI is having the most success at in 2026.
I saw some clearly AI slop made videos and got much better understanding of topics, since they don't do foreshadowing for an awe-inspiring connection that we'd admire together after 15 min videos, the clankers just get to the point. Not claiming those videos are too useful either, but I did however see accelerated learning gains from asking LLMs to explain (and do Bret Victor-style small interactive demos) math and algos, something I have never experienced from any of the educational STEM Youtube channels.
Uh huh, and what if the videos make me more generally interested in the topic? How would you measure that on a test?
Your view seems needlessly pessimistic and myopic. Just because something might not increase a number on a test that doesn't make it pointless.
EDIT: And don't get me wrong, I hope it doesn't happen for a long time, ideally never. But what would his arguments be if it does?
There ought to be people watching his channel who wonder whether given all these AI advances math is even worth studying now.
Instead, people who sell shovels win big.