Hacker News Reader

Comment by sintara | Hacker News Reader

I really struggle to understand this discussion. Mathematics isn't a way to keep a specific kind of intellectual happy by providing fun, meaningless puzzles to solve in isolation. It's the bedrock of all other science. The goal is to find solutions and to advance the field - but we seem to be treating it like some sort of occupational therapy for bored geniuses.

As we solve lower level problems, we can unlock real world math solutions. Stellerator Design, Quantum error correction, Predictive superconductivity, Inverse protein and enzyme design - the list goes on. Many solutions AI can find can bring us closer to these - so let's fire up whatever datacenter we can find and get us to these outcomes.

View on HN

Second order effects, that rational types fail to grasp

Gamedevs can make better games with AI! Yet the creatives don't get into gamedev if they didn't have to spend any effort to make their art

Mathematicians can advance mathematics with AI! Yet mathematicians don't get into mathematics because they want to "advance" mathematics

Teachers can teach better with AI, writers can write better with AI.. Can they?

Can't you people understand romance and the beauty of the human endeavor

  Yet the creatives don't get into gamedev if they didn't have to spend any effort to make their art

The art will just change. The bar will be raised. What is considered as good art in 2021 is different in 2026.

In math, the speed and innovation will increase. The bar will be raised. Those mathematicians who do not want to be part of this new world will quit and go do something else, just like how some software devs have done so since AI agents. But not all will quit.

I'm actually really excited about AI driven Gamedev. Imagine the next generation of Skyrim where rather than picking prewritten dialog options you can have natural conversations with NPCs.

Where rather than the - necessarily limited - set of predesigned outfit/equipment/face combinations each characters appearance can adjust to what happened to them. Their outfit changes with their commercial status, fights leave scars, having experienced tragedy leaves a haunted look.

Where rather than premodelled dungeons the world shapes around me to tell a story.

Gamedesign would change - and game artists would change from making 3D meshes, textures and animations to being authors of storylines and styles.

Having experienced LLM driven gamedev, I watch the movement with utter disdain. I recommend you to do the same, to actually play the products of such endeavor that you bought with your own money, for more than a dozen minutes.

> Imagine the next generation of Skyrim where rather than picking prewritten dialog options you can have natural conversations with NPCs

And of course those dialogs would be uploaded to the cloud and sold. Can't wait

Your posts here are being uploaded to the cloud and sold. Yet, you still write here. So people will still play Skyrim.

Sure, but this is clearly a public forum with the intention of communicating publicly with others. I don't know if people would necessarily be OK with the kind of speech they use during games to be treated the same.

Someone already did this, and it was a dancing bear of a thing. Cool that it worked, pretty boring in practice. The sims-like Inzoi was built around this idea as well and the blandness really put people off.

I'd love to see it happen, but until I see an AI write a compelling story worth reading, I'm glad people are still doing the heavy lifting.

Part of it is putting them in the right world and harness; I've been working on implementing bots driven by models in a 20 year old MMO with a relatively simple interface, and just a few weeks into developing the bot/model features and contracts, I'm already seeing glimpses of interesting emergent behavior in model driven worlds; alliances form, hostile players get exiled, and teams try to support each other.

From here, it's a matter of dialing in what's "the bot" vs what's "the model", then working through how to integrate them with human players. Obviously I don't have any guarantees, but it does seem like some interesting gameplay could come from it. Having a real sandbox style world with multiple clearly quantifiable goals that are at odds with each other seems to be a key way to push the models towards interesting, emergent behavior.

Also, if you haven't seen it, check out the Emergent Gardens youtube channel; it's one of my favorites, and they've been exploring these ideas for a while.

I've been playing around with this stuff for around a year too, mixing deterministic-ish simulation with AI comms, decision making and story guidance. There are flashes of cool, but the work to guide it to write good stories with what you have often takes just as much effort as writing the story. Fine, you've written a situation generator. Now you have to tell it what kind of situations are compelling, without them feeling like cookie-cutter templates. Then what you're doing is writing.

Thanks for the Emergent Garden rec, looks interesting.

You don't need to wait for the LLM driven NPCs. It already exists and has been done for years. It's always, always, always absolute dog shit and doesn't add anything to the game. Voice acting is whatever trash the LLM puts out. You get the same roundabout sentences that LLMs put out.

Character customisation already exists. AI won't make blending textures easier or harder.

>game artists would change from making 3D meshes, textures and animations to being authors of storylines and styles.

You fundamentally misunderstand game artists if you think that the meshes, textures and animations they make _today_ isn't driven by a storyline and styling. That's the absolute core of everything artists do. Otherwise we'd have reusable animation packs everywhere, like every default UE5 game does and nobody would redo a 50th running animation.

If this comes to pass, it will destroy one of my favorite aspects of gaming: talking about shared experiences. If my gaming experience is randomly generated and yours is too, there's almost nothing we share.

> Imagine the next generation of Skyrim where rather than picking prewritten dialog options you can have natural conversations with NPCs.

Back when Animal Crossing: Wild World was announced on the DS, and I heard that you could send letters to your neighbours and they would respond, naïve 11-year-old me was hoping for something along those lines.

Of course, it didn’t even come close.

Problem: they will need to have access to huge Machine Learning and inference hardware for maths specific models.

It won't be for all mathematicians, only those willing to deal with the "recipies" of ML, and those "granted" access to such hardware because a not very common resource like classic super-computers. I suspect if all that is true, maths will need bigger teams per "problem", aka will need to team up more around some math stuff.

> Gamedevs can make better games with AI! Yet the creatives don't get into gamedev if they didn't have to spend any effort to make their art

Ain’t that the truth… I’ve been slowly realizing I care very little for development when its done by AI. My main hobby, the one thing I’ve been doing since I was 13, is now as unsatisfying as it can be. Even aiming for more complicated projects I couldn’t have done before doesn’t feel satisfying

It’s satisfying when your neurons make new connections. When you work with AI they don’t.

When you work with AI, it gives you the solution. You use your brain to manage the AI. Which is not as challenging and satisfying. When AI tells you the solution and you read (skim) it, you don’t really learn or understand. You still need time and effort for your brain to make the connections.

Stephen King on writing; “Look- here’s a table covered with red cloth. On it is a cage the size of a small fish aquarium. In the cage is a white rabbit with a pink nose and pink-rimmed eyes. […] On its back, clearly marked in blue ink, is the numeral 8. […] The most interesting thing here isn’t even the carrot-munching rabbit in the cage, but the number on its back. Not a six, not a four, not nineteen-point-five. It’s an eight. This is what we’re looking at, and we all see it. I didn’t tell you. You didn’t ask me. I never opened my mouth and you never opened yours. We’re not even in the same year together, let alone the same room… except we are together. We are close. We’re having a meeting of the minds. […] We’ve engaged in an act of telepathy. No mythy-mountain shit; real telepathy.”

Is human endeavor really just "neurons" and "reward pathways", skinner boxes nonsense? A writer writes, hoping his writings reaches the mind of others. A painter paints, hoping their paintings communicate a thousand words from his spirit's depths. I think the act of making effort is more than mastery, it is a commune with oneself, and others.

You can definitely use it as an incredibly powerful learning tool. I've been using it to guide me through Haskell and I've learned so much more than I did in college when I had spent way more time on it (that time being 10 years ago so I had forgotten most of it). It's invaluable being able to go and ask the AI questions and have it rephrase the parts you don't understand. I spent a good hour with it figuring out exactly what the traverse function does (`traverse (\x -> ['z', x]) "ab"` probably doesn't do what you think it should).

Spoiler: The <> combines lists as cross joins (for-loop within another for-loop). The lambda maps to new lists ['z', 'a'] and ['z', 'b'] (strings are just char arrays). It then computes the various combinations to recombine them so ['z', 'z'], ['z', 'b'], ['a', 'z'], ['a', 'b']. Note it's not all permutations since you don't have ['z', 'a'] even though that's actually one of the "base" arrays we got from the lambda. It took me a long time to really internalize that and it's still probably not a function I'd reach for.

It is kind of cool, though, that you can do `(,) <$> lista <> listb` and you get a nested for loop.

I've been getting the most value out of LLMs this way as well. I don't care to vomit out code for the fastest MVP. I want to learn and understand what goes into what I'm building, piece by piece. So I use LLMs as a research and learning assistant to speed up gaining those tidbits. But I still have to understand the code well enough to put the pieces together by hand.

No man everything is about money and business. You don't need to be happy if your KPIs go up what are you talking about.

These people are robots.

No one stops you from doing mathematics for the romance of it, just as no one stops you from growing your own tomatoes. But if you want to grow tomatoes and make your living that way, it seems unlikely you'll make it if you go at it the same way people did 500 years ago.

Put another way - if you want your hobby to support you financially, it has to offer enough value for others, even if that value is "I enjoy knowing someone is doing this, without me benefiting directly". Simple as. Otherwise, accept it's just a hobby.

It's funny because I remember people said these things about gaming and computers when they were at their infancy

You are right that there will be some people who cannot financially continue with Mathematics as a career. I think that is horrible for society.

> I think that is horrible for society.

Why do you think that?

1. Society has (implicitly) invested resources in order get these individuals to a place where they can push and prod at the boundaries of mathematics. This investment is lost.

2. The majority of mathematicians are passionate about what they do, which typically means higher productivity. Forcibly transferring them to a different field means we lose that productivity.

3. Aspiring & passionate mathematicians will see how the field is impacted and will end up doing something else instead. This is not just a loss for the field, but also a loss for humanity because that person could have ended up being the next Terrence Tao.

Taken together, it is hard to foresee what the 2nd and 3rd order effects will be, especially because of how pure mathematics is a “trickle down” field where advancements find practical usecases later on in the sciences and in engineering.

I think this is a really meaningful insight. There's passion projects and there's - let's call it "industrial production".

Developers long had the privilege that what they were passionate about also aligned with what industries needed - at least somewhat. Then, outsourcing dev shops and later Claude and Codex turned code into industrial production. That doesn't mean we don't get to enjoy sipping a nice glass of wine and coding the 2D platformer we always wanted to exist - but it does mean that industrial production of code is less satisfying now.

The same is true for mathematicians and will be true for ever-increasing swaths of life. It's understandable to be sad about it and to miss the times when you were the mythical 10x coder with the super-high bus factor - but it is important to distinguish between activity oriented towards individual satisfaction and activity oriented towards outcomes, and not to allow the first to hinder the second.

Completely right. I think what is suprising about this wave of automation is that until now automation happened to more manual, blue collar work. I for one didn't really think that it will happen to software development up until some months ago.

This advice is so contrary to good outcomes that I am astonished. Even aside from the amazing utility of funding super smart people to follow their curiosity about the world, maths, species, string theory, neural networks, etc., just keeping it to coding, you don’t want engineering to just implement specs, you want curious people to dig under what is being said; the space of potential solutions in software is so great that you can’t waterfall your way into the best solution; it is a search, and one where the quest for elegance and abstraction and beauty enable not a ten X coder but a solution that catapults the company 100 or 1000 fold over the competitors with less elegant solutions. It’s the whole ideal behind Paul Graham’s beating the odds. It’s the story of companies that grew 70% a month while their competitors tried to make plans. Maybe you all are too young but companies that innovate and constantly delight the community are possible and systems that implement the software can be amazingly more elegant and expandable than expected. on the outsourcing, the places I have seen outsourcing succeed it is because the company can find and pay more minds either that precious gift of abstraction and simplicity and created or renovated systems so well that the outcomes based teams were just scratching their heads in wonder, if they had enough wit.

I have seen inelegant minds generate big balls of crap with AI and I have seen people wield them like scalpers thru the space of possible solutions.

I don’t disagree with using LLMs, but some puritan call for industrial efficiency as the answer is really missing the story on the last forty years of software innovation.

If the parent and grand parent represent common attitudes, no wonder so many companies are constrained by ugly software.

I really sympathize - but the truth also is that the vast majority of coding is commodity software production. "Process data from a website form" sort of stuff. I take your point that some coding is only possible when great minds go on open ended searches for novel approaches, but there's a limited subset of projects where that's a necessity.

And all that needs automation, whether by LLM or something else. Software done right should always be solving novel problems.

I really can't get behind this. The idea that our jobs being interesting and fulfilling was just a happy little accident, a weird artifact of capitalism, and that if that goes away, oh well, too bad. Back to the thankless back-breaking grind. It's just the way of the world.

That's just so defeatist and pessimistic.

It's just how the world works, for now. Maybe (and that's a big maybe) you can change it. But given the Lindy effect I wouldn't bet on it.

> But if you want to grow tomatoes and make your living that way,

You forget that we live in the age of AI abundance, "make your living" and "support you financially" aren't arguments with any weight here. Let's get clear about this, because the AI bros did a bait and switch and replaced the promised "AI abundance for you" with "sacrifice yourself for AI".

It's a really important point - math and logic skills are essential for not taking the bait. In other words, besides a science tool and an art, math is a tree where geniuses and not-so-geniuses can sharpen their claws, because being dull is an existential hazard, especially when a bunch of sneaky AI bros are hell-bent on using AI against you.

> if you want your hobby to support you financially, it has to offer enough value for others

Nah, AI is supposed to offer more than enough value for others, it's easy to do, those who can't make it happen should get out of the way, it's that simple. The BS has to stop, it's either deliver AI abundance or get rid of any and all AI altogether. There's no middle ground here, or rather, the middle ground is a swamp of existential hazards.

I think we need to separate the academic institution of mathematics from the practice of math as a human passion. When it comes to people working on math problems out of a sense of fun and curiosity AI doesn't really block them. People who are interested trying to understand math and even crack unsolved problems will always exist.

The question is if our academic institutions are able to properly support such people and enable them to do the best that they can. AI destabilizes the way we've operated in the past. So academia has to change a lot if it hopes to continue to be a place that fosters mathematical talent.

You and the parent comment are both true enough.

Somebody posted Profession by Isaac Asimov on HN a few months ago, and it really resonated with me. Your comments reminded me of it again, so I'll repost:

https://www.inf.ufpr.br/renato/profession.html

I've never read this before, it resonated with me a lot. Thank you for the share.

What you describe is a social and psychological problem.

The problem is real; the solution neither should be or could be, therefore, erect artificial barriers.

The solution is non-obvious. Historically romance and beauty have by necessity relocated to other endavors.

Which of those might remain chiefly or practically "human" is unclear. Perhaps none will, on a long enough timeline.

In which case we will either adapt—socially and psychologically—or we will rage, decline, and likely vanish.

Personally I suspect our evolved restlessness and drive to expan and explore, literally and figuratively, will ultimately need to be to our refuge. The other option appears to become lotus-eaters or trogdolytes, in the classic mold.

> It's the bedrock of all other science. The goal is to find solutions and to advance the field…

That may be true but that may not be the feature that attracts people to the field.

> As we solve lower level problems, we can unlock real world math solutions.

That is an argument often made for funding mathematics research. But again it may not be the thing that attracted people to the field.

Note I don’t share all the doom and gloom of some folks in the math community but it is a valid point that people who might otherwise have gone into the field may choose not to anymore.

Also, math research has typically focused a large amount of energy abstracting and generalizing results. Your suggestion about solving real problems usually moves in the other direction. If the powers that be with all the money decide AI can solve all the real problems that exist (rightly or wrongly) then mathematics research community will have a problem.

> Note I don’t share all the doom and gloom of some folks in the math community but it is a valid point that people who might otherwise have gone into the field may choose not to anymore.

Isn’t that only natural? I probably would have stayed in physics and not gone into software if AI had been around 25 years ago. People gravitate to areas where their innate abilities are appreciated and can make a difference.

But you also have to take into account that people from outside academia can also contribute to math. Especially nowadays with the help of AI. Just because the role of academia is shrinking doesn't mean math progress is slowing.

Math envisions and communicates logic-based structures and objects that, as a side effect, may be relevant to certain real world problems decades to centuries after the initial research.

I'd argue your statement is treating this discovery process as an engineering exercise: there exists an unexplored bounded space of logical axioms within which we want to solve the relevant problems and paradoxes. Yes, AI will help under such a framework, by giving mindless and rote repetition to brute force a solution.

But mathematics begins from a different set of primitives - the space of exploration is not guaranteed to be bounded, so the solution context is derived after a lot of trial and error. This is a different philosophical problem being solved.

I'm not sure I'm seeing that the AI solutions offered up by frontier level AI are exploring the same space as mathematics. Overlaps, certainly. But "find new and interesting problems for the sake of fun, beauty, competitive and collaborative prestige" isn't motivational to an AI.

Let me help you by breaking down your point.

> The goal is to find solutions

> and to advance the field

Nobody seems to be disputing LLMs help with the former, but some are concerned that the latter is jeopardized by having aliens do our homework.

It’s too early to tell but math could be an AI only field with automatic research and advancement.

You still need humans to tell the AI what is important.

Importance is a function of culture and biology. We already gave AI access to culture, so it just needs to capture some wild mathematicians for vivisection. Once it has a good model of mathematician biology they can be efficiently recycled.

This might be key, I think Grant Sanderson of 3blue1brown fame gave the example of the Turing machine being this small lemma that was introduced in a paper about computability, and if no one actually spent any time thinking about what it was and what it meant, we have solved the paper and thought there wasn't any interesting or useful application of this math.

Thankfully the machine was built, I do wonder if how we're going to solve the understanding part of this, there's a lot coming out, drinking from the firehose is not something I can reconcile with my experience that a lot of good ideas need serious contemplation time to bear fruit.

Eh, do you? And more importantly, will you?

"What's important" can be generic stuff like "fusion", and current gen LLMs are already more than able to go from there and iteratively explore what they might contribute with. It's not exactly a stretch to take the extra step of asking an LLM about what the top 10 unsolved scientific problems are.

I say this because I see this mantra repeated often, but largely it makes very little sense. Do you have any concrete examples of unsolved important problems that an LLM is inherently unable to locate?

'Figuring out what is important/what success looks like' in math and other fields seems much easier for humans than 'achieving important success', which is why we celebrate success.

Why do you think AI will be better at achieving success than humans, but worse than humans at identifying it?

Unless you mean super high level stuff like prompting Galactica 8.5 with "reorganize human society for maximum flourishing, make no mistakes"

I worked on formal proof in 2014, but abandoned it when I realized academic mathematicians would never use it. Formal proof would have helped math’s consumers: better search, fewer mistakes, easier translation (to other human languages), etc.. But academic mathematicians didn’t care about math’s users very much.

Now that there are tools that help math’s consumers and because the economy of proof production wasn’t oriented towards consumers, there is a real chance that academic mathematicians get sidelined.

The field before AI was build up so that humans can learn and understand and teach other humans how to advance these fields.

AI is here 'suddenly'.

Let them discuss this out, let them go through the emotions.

And its unclear today IF we still need anyone doing anything in 5 years or not. In theory we could also just freeze 90% of things we do and put everythig into AI as a human race and just let AI do everything in 5 years.

We are not doing this because we don't know yet if it will play out like this, we don't know how much a human still needs to understand.

And to be clear, what you see as a 'problem' is just a capitalistic issue, not a human race issue: Its absolutly fine for us as humans to persue understanding for the sake of it

True, but then you're into the super big picture questions that The Culture books were generally asking.

Considering the 10ft-horizon-thinking around us, I feel like this is a feature not a bug.

Yep!

And i'm looking forward to it to a certain degree. If progress is not stopping or showing any slowdown, we will run either in very basic physical limits or we will continue having quite an interesting time ahead.

I'm sorry but this can't be a serious take. Can AI chop wood in my backyard? Can it go grocery shopping for me? Can it build my garden shed? I highly doubt, in fact I disagree completely that "AI can do everything in 5 years"

I meant only AI not physical AI.

But Physical AI is on the path to be done in 30 years.

> put everythig into AI as a human race and just let AI do everything in 5 years.

That’s the SV sci-fi fantasy to World Domination via LLMs, which is actually a delusion.

There is a huge gap between an AI doing a specific task better than all humans and AIs working on all the tasks humanity needs. Humans do trillions of those tasks everyday.

Yes, this is another goalpost, but it’s nowhere in sight.

Human race is huge. Even if we focus only on compute, human race has 8 billion AGIs, each work with very little power. Yet sustaining energy (food) for human race is a real problem.

The main fallacy here is setting some variable to infinity in a back-of-the-envelope estimation. But there is no infinity in the real world and things don’t have to scale as predicted.

The gap is closing with every release. In parallel we get improvements in chips, Google announced 'dreaming' and there might just be an AGI level of thinking.

I'm not seeing any slow down right now and I would call myself also a more diverse person knowing and acting on more than just software development. I don't have the feeling that i had to learn that many extra skills to try myself on other topics.

IF the world models or LLMs learn to simulate at runtime, they can simulate a task and do adhoc RL.

With the current power structures in place, assuming AI can really do all labor, why would the people in power keep us around then? What for?

They probably wouldn't go for the mass murder option right away because of the logistics issues and possible revolts but there is no reason to keep excess population around long term.

Yes, technically, they could realize that there is enough wealth for everyone now. They will not. The problem with capitalism is that it can't stop. They will try to accumulate more and more wealth until the planet is destroyed.

We will see the future that Start Trek promised, eventually. If humanity does not get wiped out before. But what is coming first will be a wave of bloody revolutions, it will not be pretty but necessary.

Thankfully we probably have quite a bit more time than five years left, realistically.

For humanitys sake i would argue.

But yes it will be very interesting if we don't reach any ceiling if another fitness function than "more money" will arise or not.

I think the story "Manna" describes pretty well what will happen if AI and robots do all/most work. People will basically be pushed into welfare buildings and receive minimum support to survive. I don't think the capitalists will share the wealth voluntarily. If they wanted, they could do that already, but they don't.

That last value is what has unleashed the knowledge and power under the prosperity capitalism depends on.

I think "fun" and "meaningless" are important parts of math. There is applied math for sole utility. Other math is just random search that may or may not be useful in long term. But later people/AI will find important use cases from these short-term "meaningless" researches.

Mathematicians both build/formulate abstract conceptual frameworks and formulate problems within those frameworks. Solving the problems has historically been one of advanced the conceptual frameworks.

Now, what has been useful for advancing physics and biology has been the abstract conceptual frameworks - using Hilbert Space for model quantum mechanics or tropical geometry for DNA modeling.

So far AIs have shown an ability to prove mathematical theories but not an ability formulate coherent-to-human mathematical models. So allowing AIs to simply replace mathematicians isn't so far a way to achieve broad advances in other fields. Indeed, AIs so far are great mathematicians only if one views mathematics as solving "meaningless puzzles ... in isolation"

We treat pure curiosity, about maths but also about basic science of all kinds, as a self-sufficient goal for a portion of our civilizational effort, because it has paid us back handsomely in ways that some centralized planning approaches would have missed. Suppress curiosity and fall behind. Give some of those bored geniuses freedom to work on their chosen problems and win. Not to understand this is to not understand your technological society.

It's both. Some math has applications. Some math doesn't. Many mathematicians are interested in mathematical beauty, in solving problems, and playing with puzzles.

None of the results from October 6, as far as I know, have direct real-world applications today. We can't go build a better thing, or make a faster thing, etc. I might be missing something (I'm not an expert in any of the fields). I don't think most folks would have a problem with solving a bottleneck problem that let us flux a capacitor at 2x efficiency, or whatever.

I also think this is hard for a lot of folks who are in fields that heavily apply math to come to terms with. I do not think that programming and practicing mathematics are the same: many (most?) mathematicians are closer to a composer than a programmer: all three are making things, but there is much more emphasis on elegance and cleverness in mathematics than in programming.

But again, it's not binary, and a lot of the discussion does try to take a black-or-white view of it. It's tradeoffs all the way down.

For what it's worth, I think this quote from the article sums up my position best:

"To me, someone who chooses a worthwhile problem, obtains a proof completely using AI, understands it thoroughly, and uses it to develop interesting further questions or directions has made a substantial mathematical contribution. I would give that person substantial credit even if they supplied no new ingredient in the original proof. Choosing worthwhile problems may end up mattering more than producing proofs."

> Mathematics isn't a way to keep a specific kind of intellectual happy by providing fun, meaningless puzzles to solve in isolation.

Yes it is. At least in part.

> It's the bedrock of all other science.

Yes it is. At least in part.

> The goal is to find solutions and to advance the field

That is only one of the goals - but it is not the goal.

> we seem to be treating it like some sort of occupational therapy for bored geniuses.

I certainly hope not.

> Many solutions AI can find can bring us closer to these - so let's fire up whatever datacenter we can find and get us to these outcomes.

This has nothing to do with your earlier points, but absolutely - let's do that!

If I may, I think that in general, you are confusing math-as-tool with math-as-art.

Most math degrees literally give out a Bachelor's of Arts.

It is a tool at some points, art at others - and sometimes the art happened to become a really important tool.

By all means, use AI for building tools.

But you can't ignore its deep rich history as an art.

I think that math can still be a creative endeavor, ultimately. In many cases it is not so creative, and really is an exercise in searching the potential solution space (which it is now clear the AI excels at). However, math can also be about the creation (discovery?) of new language itself, which in turn enables us to ask questions that we couldn't previously ask in the first place.

As a mathematician, I don't disagree with you, I do think the recent events have exposed a certain brand of elitism/gatekeeping etc within the mathematical community. While that may be true, it's not always about "finding the solution", oftentimes we first have to learn how to even frame the problem to begin with.

> The goal is to find solutions and to advance the field

And why are we finding those solutions and advancing the field if not to improve lives? What’s the goal otherwise?

> so let's fire up whatever datacenter we can find

Because this is actively ruining our lives. This relentless, selfish drive for more and more data centres is hurting communities and the environment (you know, where people live), increasing prices of other things, consolidating wealth inequality, and much much more.

So no, let’s not “fire up whatever we can find”. How about we instead take a break from consuming vast resources, hurting people, and making everything worse, to work on the very solvable problems we have right now, which are very addressable with the things we already know.

I guess the issue is -

To advance math you need people who could be really high earners in any other field and stuck to math mainly because they find it intrinsically motivating.

You change the calculus for motivation you potentially devastate a generation of talent who just decide to go to wallstreet or something since it seems there is no longer fun to be had in maths.

It actually _is_ about individual motivation.

AI:s are not mature enough to have any sort of institutional framework built around them.

So removing a generation of mathematicians is institutionally destructive and since the field seems to have tacit elements - you teacher and the student to discuss things in same room - this is not just about labour but continuity of capability.

I mean that's how I understand the alarm. I don't do maths so I can't argue for or against.

AI:s are amnesiacs. Even with 1M context they struggle really hard to understand architecture of complex software application.

So you can't really use them as only cognitive building blocks for complex novel systems design and engineering.

I believe the same analogue probably applies to maths.

So that is the way you want to live ? Just being mechanical ? I don't see the point of solving those problems if it mean that our day to day life is shit. The point of tech is to make our life easier, AI doesn't. It just want to replace us as it is today

Craftsmen of all kind probably said the same thing when machines suddenly did their work cheaper and faster (and better?).

>so let's fire up whatever datacenter we can find and get us to these outcomes

You're delusional if you think this is what these datacenters are built for. The goal of AI is commercialization, not advancing the human race. Sure, both can be done in parallel, but dont be fooled, immediate money is worth more to stakeholders than advancements with results nobody can turn into money.

This seems obvious in current year, but extrapolating it I wonder where it takes us.

If society isn't there to keep us safe, free, and happy - what's the point? Accumulating lean certificates and energy production capacity?

Well, I think the argument is that the average person gets safer and happier when you stop letting dorks scam their way out of 12 hour shifts the hospital / factory

Terry Tao is a child prodigy who's entire life has been 'you are so good at math' is it a surprise it's a bit reality shattering to him to find an AI can write proofs better than most humans?

You missed the entire point. He’s looking for a way forward. This post wasn’t written by him and doesn’t discuss being shattered at all.

> It's the bedrock of all other science.

no, its just a language. physics constrains 1+1. "one plus one equals two" is difficult to transform

Good luck trying to do science without a language.

> > It's the bedrock of all other science.

> no, its just a language

Totally disagree. Math and physics overlap, for example, but the idea of a conservation law that your system follows is way more than language. Math provides a unified view of all of the systems with conservation laws. That’s an idea, not just vocabulary.

> Totally disagree.

instead of turning where measurement ends and language begins into a debate on social media

go outside and count some rocks for a while

use whatever symbols u like

return only when the symbols change after recounting

restricting the start of the debate between u and the rocks is healthier and more peaceful

Reply on news.ycombinator.com