Leetcode does not test your ability to code in the sense of anything related to real world engineering applications.
Leetcode merely tests your ability to memorize and regurgitate algorithms from a textbook -- a skill that LLMs can now do a thousand times faster and more accurately than any human.
It never had much to do with real world software engineering, and now it's even less relevant than ever.
True, but this is also a stunning example of survivorship bias.
Countless other mathematical curiosities were developed in the 1700s -- and forgotten. Calculus just happens to be the one that found practical applications later, so that's the one that's well known today.
It's difficult to draw the conclusion that "every possible branch of learning ought to be funded" by appealing to "later practical value" based on this cherrypicked example.
On the other hand, clearly _some_ novel theoretical work with no apparent immediate value _does_ yield real world benefit later.
Since we lack the resources to fund every PhD with a crazy theory on what the next new subfield ought to be, how do we decide?
Seems like AI could help massively there, by removing a huge bottleneck around technical elaboration and application seeking.
> Countless other mathematical curiosities were developed in the 1700s -- and forgotten. Calculus just happens to be the one that found practical applications later, so that's the one that's well known today.
1. You may not agree, but some people will argue that knowledge is worth accumulating in itself. We fund astronomy well beyond the solar system despite there being no real prospect of practical applications.
2. The value of calculus justifies the cost of all the curiosities. The economic benefits of funding lots of "curiosities" was well worth the few that were useful.
1. is debatable, but in any case that is not the argument that the GP post made -- the post specifically argued that it's worth developing knowledge with no immediate known application, because some application may be developed later.
That's a different argument than "knowledge ought to be accumulated for its own sake."
2. is unclear; the economic funding model of the 18th century was very different. At that time, there were very few academics, and nearly all were independently wealthy and self-funding. That is, they chose to spend their own money on mathematical curiosities to amuse themselves.
Now, in modern academia, we are discussing the case that public tax money ought to be appropriated and allocated to pure knowledge / curiosity discovery _at scale_, for work which -- by definition -- has no known application or use. This is a very different economic and ethical proposal. This is no longer the question of how one rich man chooses to spend his own money; we are discussing the social application of common resources -- compulsory tax payments for all.
It's easy to imagine other ways to spend that tax money that have much less debatable social value -- how about free healthcare for all? Better roads, or high speed trains? No famines? Flood control? There are a million other ideas with immediate benefit.
I appreciate the value of pure knowledge as much as anyone, but it's also hard to make the case that it should be pursued indefinitely, whereever possible, and without limit or boundary when so many competing priorities exist. It's a question of competition for very scarce resources, versus unlimited wants. How do we decide appropriate allocations?
in general, yes. but specifically studying exoplant A or exoplanet B? or understanding dark matter? or driving a little science robot on Mars? what's the immediate utility of that?
The same thing could have been asked of radio waves when they were first discovered. It was actually considered purely an academic curiosity with no practical application. You can't know ahead of time which discoveries will be the lynchpin for some as yet unimagined future advance. Much like VC you need to spread out your bets and expect that most won't make direct return, but that every now and then something with 10000x return will surface.
right, which is all the more reason to continue studying them in spite of the lack of immediate returns. our capitalist mode of quarter-driven roadmaps means we no longer care about the long horizon, it's all 'results this, metrics that, now now now' - meanwhile any student of history who's ever looked at the history of computing, or data science, or virtually any of our modern marvels knows how much purely abstract stone had to first be cut to get us to the results we have today
can you imagine chucking all of computer science away because Babbage's thinking machines and Lovelace's programming cards didn't really result in meaningful capital gains here in FY26Q3? or, a more realistic example - imagine biology and genetics today if Mendel's discoveries had been published in a more widely read journal, if he hadn't been outright dismissed for his mathematical approach. or Semmelweis' advocacy of antispetic procedures that flew in the face of the then humors-based understanding of medicine in the 1800s - how many lives would have been saved between then and when Germ Theory finally took off decades later?
it's ironic to me that it's often the same people who'll adopt the logic of banning reproductive rights on the off chance that someone very unprepared to have a child might birth someone who can cure cancer in spite of the probabilistic evidence of SES being highly correlated with educational attainment (eg that graph of the income bracket that the parents of all Nobel Prize winners start with). and yet when it comes to the actual nascent research and development required to pave the way for larger discoveries, that's seen as totally pointless and not worthy of funding. it's truly flabbergasting
Agreed. For a long time it felt like most governments/funders accepted that pure research is a necessary part of feeding the applied research machine. And now the funders seem particularly short-sighted.
In saying that, in the spirit of a good faith argument, is there value in temporarily deprioritising pure research in (current) times of funding drought?
> 2. The value of calculus justifies the cost of all the curiosities. The economic benefits of funding lots of "curiosities" was well worth the few that were useful.
That is totally not true. I am sure you can hardly name even one of those "curiosities" that later turn out to be useful and not "free" (i.e. if not existed, wouldn't be invented on the spot as a tool for solving those particular problem that they end up solving).
> this is also a stunning example of survivorship bias.
No, it's an example of why you have to allow people to pursue what at the time look like "curiosities", even though most of them don't go anywhere--because the very small portion that do go somewhere, end up changing the world, and we don't know in advance which ones those are going to be.
It's quite true that the funds we have for this are a finite resource. But that doesn't mean that "foreseeable practical applications" is a useful filter for how to deploy that resource.
Yes you have to allow people to pursue these things. That's not a compelling argument for funding, especially when nobody was funding the people that did invent calculus.
> nobody was funding the people that did invent calculus.
Of course someone was. Newton and Leibniz weren't living in autarky an some island. They had an income stream that was taken away from somewhere else and given to them so that they could afford working on their pet projects.
The fact that it's not publicly funded doesn't change the fact that there's still resource allocation happening.
> every mathematician in the current day is funded,
This is tautological, “mathematician” being defined in the modern world as someone paid to do mathematics…
Notice that the modern world is able able to sustain significantly more mathematicians than when “mathematicians” used to mean “someone well-off enough to work on mathematics in their ample amount of free time without worrying about their subsistence”.
> This is tautological, “mathematician” being defined in the modern world as someone paid to do mathematics…
Note that I'm including all the ones that a normal definition would not call paid, but your definition does.
But yes when a definition declares that everyone doing a thing is "funded", it does get pretty tautological and get in the way of arguments for actual funding.
Newton and Leibniz both inherited substantial estates from their upper middle class parents.
Neither was fabulously wealthy, but they were both able to bootstrap their academic careers with inherited money until they became famous enough to receive patronage.
"Individuals choosing voluntarily to spend their own money researching mathematical puzzles and curiosities and pure knowledge" is a very different ethical proposition than modern academia, which depends heavily on compulsory tax payments levied on everyone.
During the years when Newton made his great discoveries, of calculus and universal gravitation, he had a scholarship from the University of Cambridge. (Note that it was not until a couple of decades later that Newton published this work.) After he got his MA he became a Fellow of Trinity College. So society was basically subsidizing him.
Newton was able to attend Cambridge at all only because his parents were prosperous yeoman farmers who owned substantial agricultural estates and held private incomes.
Nearly all schools at the time were not subsidized by society and required substantial tuition payments. Most children had to get jobs. School was rare.
Newton's parents, while not extremely wealthy, could afford to send him to expensive private schools, instead of requiring him to get a job at age 8. This was a prerequisite for entering Cambridge.
His early career was subsidized by his upper middle class parents and inherited capital, not by society at large. He became a professor at age 26, and was paid a salary in exchange for work after that.
You know how drunken debates with friends over trivia became pointless with the advent of Google and Wikipedia. The debates didn't get better. But they could. They did. Today we have HN mods and HN citations and upvotes
>useful filter
There's also "useful friction" as _almost_ imagined by Tao here
That's an empirical question. And you also need to justify why your country in particular should finance that.
For the kind of research where benefits accrue to the inventor (or her employer) and others can be excluded so that the benefits don't 'spill over', then private companies can fund it.
For the kind of research that spills over, you can just let the tax payers of that other country foot the bill. Eg the US can free-ride on Chinese research, and if having mathematicians in the population is good, the US can offer green cards to whatever has a math degree from a good enough university.
There might be some intermediate research that has just enough spill over that a company won't do it, but a country might, should be a rare creature. And differently sized countries should have different sweet spots: some companies are bigger than some countries after all. But I don't think we see different countries select the research (mathematical or otherwise) with an eye towards exactly tailoring spill over.
You can call it survivorship bias, but another way to phrase it is that it's very difficult to forecast the practical benefit of any one piece of mathematics even while the long-term impact of mathematics as a whole is undeniable. And any schemes to further ration resources among mathematicians ignores the reality that for such foundational subject, math research already one of the least funded compared to other disciplines or domestic priorities.
"Falsifiability" is generally understood in terms of physical observations. Mathematics is ultimately tautological: they are true or false by their own definition, without reference to the physical world.
It just so happens that certain kinds of mathematics are unreasonably effective in drawing parallels to the physical world, but as far as mathematicians are concerned those mathematics are neither better nor worse than those that do not correspond to anything tangible.
> those mathematics are neither better nor worse than those that do not correspond to anything tangible.
A sweeping generalization. I certainly know professional mathematicians who disagree.
Also: many consider “inter-connectedness”, not “tangible” to be a sign that a topic is interesting. That is, it touches branches of mathematics aside from its own.
> Seems like AI could help massively there, by removing a huge bottleneck around technical elaboration and application seeking.
Are you proposing that humanity should rely on AI to determine the fields of study that should be perused and those that should be defunded?
Why don't you first come up with an AI that can predict if the stock market will go up or down tomorrow, with 99.999% accuracy. Should be really simple as it only needs to answer what will happen tomorrow.
After that, come up with the AI that will predict how actions or non-actions today may affect outcomes 100 years into the future.
> Are you proposing that humanity should rely on AI to determine the fields of study that should be perused and those that should be defunded?
We already have examples of politicians relying on chatbots to understand problems. I wouldn't be at all surprised if, given a few more years, the above will be effectively the case without any deliberate effort.
The sad thing is that I would actually trust a chatbot more. Not because they are that good, but because the bar for good governance is that low, especially of late.
And if calculus was the only useful thing to come out of 1600s mathematical research, it would have been worth it. The other dead ends don't need to justify themselves. Getting one thing of this magnitude justifies it all
On how to select what to fund in the future: find the brightest minds, fund whatever they want to do. That tends to work out in aggregate
> And if calculus was the only useful thing to come out of 1600s mathematical research, it would have been worth it. The other dead ends don't need to justify themselves. Getting one thing of this magnitude justifies it all
It would have been worth what? Doing mathematical research at all? Nobody is suggesting we shut down mathematical research.
It doesn't justify a funding system that didn't exist / didn't fund those researchers. We need to come up with better reasons to fund such a thing.
> It doesn't justify a funding system that didn't exist / didn't fund those researchers. We need to come up with better reasons to fund such a thing.
Do we?
For most of us, it seems like the EV (expected value) from such funding is quite positive. The work pays for itself in the long run. That's justification enough.
> But I would suggest replacing "seems" with the best numbers available.
I know there's research on it for applied and basic science in general, and the rewards were massively EV+. But I can't recall whether similar research covers applied and basic mathematics research.
I actually just think every person should have their basic survival provided for efficiently by society. My argument is that most people only veg out because they are tired of a hectic work life, but if given the chance to coast for a long time they would find productive work voluntarily and be more inclined to share their work. My claim then, is that society would net a benefit from this arrangement which would more than pay for itself. And you don’t have to try to pick winners and losers.
> My argument is that most people only veg out because they are tired of a hectic work life, but if given the chance to coast for a long time they would find productive work voluntarily and be more inclined to share their work
We have natural experiments of this: retirees, both early and otherwise. Do retirees during the first five years of their retirement "find productive work voluntarily" (your choice of words, not mine) in quantities that are comparable to what they did at work during the five years prior to their retirement? Not exceptional individuals, but on average.
I retired circa five years ago and spend my time doing stuff like walking the dog, drinking coffee and reading books, much like other retirees. My productivity is as close to zero as I can muster, and it is the rule, not the exception.
There's an obvious bias in focusing on retirees though. I mean, the whole notion of retirement exists for a reason.
But also, in all honesty, if we can feed and shelter everyone on the planet with only a small proportion of the population engaged in productive labor, what's the problem with most people walking the dog, drinking coffee, and reading books?
> There's an obvious bias in focusing on retirees though. I mean, the whole notion of retirement exists for a reason.
Go on, please. Whatever argument you make, please account for early retirees as well.
> if we can feed and shelter everyone on the planet with only a small proportion of the population engaged in productive labor
But that is not the premise of the discussion at hand, though. My comment was written in this specific context:
>>> My argument is that most people only veg out because they are tired of a hectic work life, but if given the chance to coast for a long time they would find productive work voluntarily
I think everyone deserves the opportunity. Most will still veg out even if they had no minimum wage blue collar monotonous job. But everyone deserves a chance. Some people are born into it and have no chance. While I was lucky to have a supportive financially stable family through no work of my own yet I wasted it all scrolling my phone.
It's very unlikely anyone with the intelligence to contribute to math research can't already find a non-hectic job that provides for their basic survival. People choose hectic jobs because they want to do better than basic survival.
This is such an insane take I don’t even know where to start. Even in the western world people increasingly need to work more than one job / have multiple incomes to live a life where basic necessities are given, and some luxuries are attainable (read: vacation, not rolex). That’s without children, in rich countries. That isn’t even touching on the problems someone e.g. from a lower caste in India might face. Asserting that „everyone with the intelligence to contribute to math“ would be able to just “find a job” that would them also contribute to the field is completely detached from reality.
People working multiple jobs in the western world have some reason beyond "basic survival". Supporting other people, living in a desirable location, or like the vast majority of people they want to make enough money to go beyond basic survival because basic survival is low status. Which is why the idea that all we have to do is give people enough money for a basic survival lifestyle and they'll be happy and spend the rest of their time on things with perceived aesthetic value like math research is the insane take.
You’re leaving out the niceties like living in a neighborhood where you don’t wake up and see if your child is still alive every time you hear gunshots in the night. Or working a job that doesn’t risk your limbs or even your life if you make a mistake. Or working for someone who will not just fire you the instant you stop being able to work due to an on-the-job injury.
Most of these aren’t even new inventions; read about the working conditions in the 19th century.
> I actually just think every person should have their basic survival provided for efficiently by society
Define basic survival. Tantalum mine worker in the Congo who earns 22 cents per hour and raises his three children on that money, is he under "basic survival" level or below?
And to be clear, as a part of a society, are you ready to forgo your salary, except maybe 1 dollar per hour, to provide to other people? Why haven’t you done it yet?
I think this mostly comes down to what work is productive for a person. People who are fascinated by research topics are likely to work on them and people who are enamoured with creating art will do so. But people who's productivity was largely being some cog in some machine will be all too happy to say good riddance.
Also it ignores the fact that the core of calculus and the idea of dealing with infinitesimal quantities was independently discovered for many centuries. eg. Archimedes use techniques that were eerily similar to integrals.
The work that Newton and Leibniz did was in formalizing it and coming up with the notation that would end up being more widely accepted and applicable. It's very likely that the techniques would have been developed later to solve a practical problem.
You bemoan the cherry picked example and counter with an unfalsifiable claim. Certainly we have remembered much more math than Calculus, and much of it has been of practical use.
How can we hope to quantity the expenditure on math we've collectively forgotten? It's unknowable by definition. The only reasonable thing to do is to determine the value added after the expense paid. Even in a world where calculus is the only thing that we took away from the math of 1700s my guess is that this is still an economically beneficial calculation.
The claim is falsifiable; the bulk of 18th century math is "forgotten" in the sense that few, if any, people still use or apply or even know about it.
It's not "forgotten" in the technical sense that one _can_ still go dig into the dusty archives of any number of old university libraries, and review learned journals, diaries, commonplace books, personal correspondence, and so on from the 1700s that describe the mathematical work of the day in detail.
You can then systematically review that work, and test whether the claim that "nearly all mathematical output of the 18th century has been generally forgotten and never found any use."
I hypothesize that this experiment will show that nearly all of the mathematical output of the time long ago fell into oblivion. This is a falsifiable claim.
Your claim is still rubissh, as you neglected to interact with the rest of my comment: economic utility does not necessitate all mathematical output directly contributes.
You redefine "forgot" to make it falsiable, but also neglect that you need to refute that some "forgotten" work didn't contribute to new work down the line.
It's not survivorship bias, because math isn't a bunch of independent, parallel things, where one turned out to be useful and the rest was junk. There is no known way to advance only the portions of math that, centuries later, will turn out to be economically useful. Even with AI, we only know how to advance the entire subject.
You're effectively asking to predict the future hundreds of years in advance. Nobody and nothing can do that. With math, as with all science, you must be willing to accept that not everything will be a hit. There will be misses, and often the same person will generate both hits and misses, because it's fundamentally unpredictable what remains a miss and what doesn't over centuries. The best demonstration of this is that your own thinking here would've banned the invention of calculus: it didn't materially affect daily life for a solid century.
> True, but this is also a stunning example of survivorship bias.
It's not “survivorship bias”, it's merely the illustration that mathematics are a strong link problem: it's the strongest result that determines the impact of the field (not the weakest, like in weak-link problems).
>> imagine yourself living in the 1700s. how would you justify Newton and Leibniz's work on calculus
The direct use of that math to biuld better weapons. That math was/is essential to the development of modern artillery. The first tasks assigned most early computers were to calculate ballistic trajectories, and also tide tables which also have immense military applications. Math was and is a weapon.
> Countless other mathematical curiosities were developed in the 1700s -- and forgotten.
Really? Like what?
edit: this is either an unfalsifiable claim, because by "forgotten" you meant here is no extant knowledge or remaining record of it, or it's almost certainly nonsense and anything you could cite would be foundational to some area of modern mathematics, even if as a disproven counter theory.
A maybe example that springs to mind is how Gauss discovered the FFT in the early 1800s (predating even Fourier analysis) but didn't find it interesting enough to publish, so while his work was important, it was also forgotten and had to be rediscovered.
"... but this margin is not large enough to contain it." Sound familiar?
Now, it's possible that Fermat thought he had a proof of his theorem, but that it would have turned out to be wrong. That happens sometimes. But he did not write it down on any document that has survived to this day, so we know that he had something that has been lost.
It's similar to history. We know that there are books, plays, etc. that used to exist, but that nobody knows today. Because other writings that have survived to this day quote from them. But beyond the quote, we don't know anything else about the play or the book or whatever.
Fermat's Last Theorem almost certainly isn't the only case of mathematical theories that we know were published somewhere but that we have no record of today. It's just the only one that I happen to know about, not being a mathematician myself. I'm sure there are some people on HN who know of others.
Really hardcore spherical trigonometry comes to mind, but I may be off by a century. It used to be considered fundamental, but almost nobody besides maybe a few historians of mathematics knows the methods anymore.
> Seems like AI could help massively there, by removing a huge bottleneck around technical elaboration and application seeking.
Ah yes, lets take the existing system that's already barely holding up and flood it with slop. What could go wrong?
> Since we lack the resources to fund every PhD with a crazy theory on what the next new subfield ought to be, how do we decide?
Do away with the perpetual uncertainty of the "will we or won't we continue to fund you" grant treadmill for practicing academics. Hold a yearly competition of academic prowess and intelligence open to any adult US citizen under retirement age. Award the winners a grant good for 40 years. If we allocated slots equal to 0.005% of the population each year (ie ~17k) that would represent 0.2% (ie 1 per 500) of the population in total at any given time.
Obviously I say that (mostly) in jest but clearly there are workable solutions if we approach things from a new angle instead of determinedly clinging to the status quo.
Not if the billionaire who has been funding the endowment for their university's math department wakss up one morning to read a three sentence Claude summary of this thread, and says "hmm... sounds like what we really need is a ChatGPT moment for math" and decides to redirect his cash to the cutting edge of next gen slop generation tech instead.
Don't despair; new tool rollouts are always a hard sell -- especially when replacing a beloved classic.
For me, the killer feature would be "ease of use."
Sure, I _can_ make rsync do any number of rarely-used optimizations, if I feel like studying the manpage for half an hour and figuring out how to fit it to my exact use case.
If your tool has the same features but is automatically adaptive -- I'd use it.
Copying thousands of tiny files in deeply nested subdirs? Just works.
Copying a huge file that's already encrypted? Just works.
Copying a mix? Just works.
No special flags to set, zero config. It just works, optimally, every time.
"Reducing the user's cognitive load" is the killer feature here.
You believe you know better than the authors themselves about how they ought to be treated?
They chose -- explicitly, voluntarily, knowingly -- to give the thing away for free. That's their right for their own creation. Your preferences don't supersede theirs.
You've emphasized this point more than once, but I don't think it's a good counterargument to the GP. It is equally valid to say that they have explicitly (voluntarily, knowlingly) chosen to accept donations for their work.
By profiting off their work and not supporting them you are making your own choice. It says nothing about the authors' preference.
They chose a license that is _explicitly designed_ to allow people to profit from their work without paying.
There are many other licensing terms in common use. They chose that one.
That's fine, they can do what they like with their work. There is no moral obligation to reinterpret their motives in a fashion that differs from their formally expressed preferences.
They chose to accept donations -- also fine. If you want to donate, go ahead. If you don't -- that's fine too, by definition.
Surely "the right thing to do" includes some element of the authors' explicitly stated preferences about how they wish to be treated.
In this case, they chose -- explicitly -- to NOT require any payment or contribution from anyone.
Your feelings about what constitutes "the right thing to do" are apparently different from those of the authors. You believe you know better than they themselves do how they should be treated.
They clearly would like donations, but choose not to require any payments.
Your line of reasoning seems to be along the lines of "anything not forbidden is compulsory" which is not a good place. Having the choice to donate but not being required to is good.
And yes, I think that if OpenAI is choosing to include an entire copy of LibreOffice in their product, it is definitely the right thing to do to sling them a donation. Note that the moral thing to do is not always the required thing to do, or even the legal thing to do.
"It's the right thing to do" in reply to a moral query is the "begging the question fallacy".
The question at hand is of the form: Is it right to do X (to donate, in this case)? Your reply amounts to: yes, because X is the right thing to do. This merely restates the query as its own conclusion.
"Donations" are - by definition - entirely voluntary.
If people want to donate, that's fine. If they don't want to donate -- also fine. That's what "donation" means.
Therefore anyone's decision not to donate is entirely consistent with the author's own expressed preferences.
Whether an actor happens to be large and rich (OpenAI) is irrelevant. The license -- the author's formally expressed preferences -- does not say anything about "free unless you are rich." They could certainly have done so if they had wanted to. They didn't. You are taking it upon yourself to claim that you know what's best for them.
More like, don't expect anything if not forced. Most people only generally pay for things if there is no other option. It's about what's moral (even then debatable as everyone has different morals) versus what's realistic to expect.
That article has no detail. How much did they make? They said they hope to break even in 2019 so they weren't profitable, what are the latest stats? Compare this to NYTimes which has been crushing it after implementing their paywalled subscription.
> Surely "the right thing to do" includes some element of the authors' explicitly stated preferences about how they wish to be treated.
It does! And they wish to get donations. There's a button right at the top on the site making that clear.
> In this case, they chose -- explicitly -- to NOT require any payment or contribution from anyone.
Yes, so what? It is true that they wished to not require payment. That is one of their wishes, but they have more than one wish. And if they required payment, that wouldn't be a donation anyway.
So.... Since they explicitly chose not to require any payment, there is no moral obligation to pay them. That's the point.
Would it be nice? Sure. But "not donating" is also fine.
"Donations" are, by definition, entirely voluntary. Whether someone using the software under its proper licensing terms is wealthy, whether you approve of their business model, or other factors are all irrelevant.
> So.... Since they explicitly chose not to require any payment, there is no moral obligation to pay them. That's the point.
Requirements are about legal obligations, not moral ones.
> Would it be nice? Sure. But "not donating" is also fine.
It's in general fine not to donate to libreoffice. But that doesn't extend to 100% of circumstances.
> "Donations" are, by definition, entirely voluntary. Whether someone using the software under its proper licensing terms is wealthy, whether you approve of their business model, or other factors are all irrelevant.
As a third party I can judge someone for not donating based on their circumstances. It's not irrelevant. The moral implications are not part of the definition of donation.
The fact that these same people can't tell whether it was authored by a human or an AI without a seperate disclosure is strong evidence that human authorship as such is not what they value at all.
Imagine I sell you something and you pay me in cash. The notes look authentic, but I later discover that they are conterfeits. It would be absurd to claim that your successful passing-off deception demonstrates that I don't value money at all.
If you find a counterfeit bill, you experience a loss. Imagine if instead of a counterfeit bill you got a counterfeit movie. The quality is exactly like the real deal, but it's an illegal copy. You find out after the fact. Does this affect your enjoyment of the movie? You've been bamboozled either way, but in one scenario you lose out and in another scenario the people selling the movie lose out.
Whatever the morality involved, it's an unstable norm. There can be good actors who use AI and disclose, and good actors who don't use AI and don't disclose. There are bad actors, however, who will use AI and not disclose. The entire discourse will be flooded with non-disclosers accusing each other of AI use (particularly, the bad actors will accuse everyone else of AI use, to flood the zone with shit), and there will be no way for a consumer to distinguish good from bad. Bad actors will prevail economically.
See e.g. corporate green and rainbow washing, Amazon marketplace, organic food.
I agree that the current culture of disclosure is unstable. Nevertheless, I think that there will still eventually be a mechanism to verify the degree and manner to which AI was involved in the production of art, though perhaps with very different economic scale and distribution channels. Speculating, perhaps there may be an "organic food" niche of mass media art.
On the contrary; it seems to me that caring about its authorship even though the product may be outwardly indistinguishable from something a human could have made shows that human authorship is, in fact, precisely what they value.
Are you trying to make an argument that "if they really valued human authorship so highly, they'd devote their every waking moment to learning what every possible method is to tell when a given work was created by an LLM" or something? Because I would think that was fairly obviously bullshit by this point. For one thing, there's plenty of ways to make LLM-generated works (especially smaller, simpler ones) that are 100% indistinguishable from something human-created. (This is especially true in text formats, where the amount of information, and thus the possible failure modes, available in the output is smaller than in images or video.)
Art consists of decisions made by the artist that are interpretable by the audience. Automatically generated imagery lacks this, but still presents a fascimille of it. The intention and thoughtfulness of the artist is encoded within their creation, and when there is no artist this is lost.
Just because it takes effort or time to tell a fake doesn't mean its not important.
If you buy a fake that looks extremely close but falls apart a month later, that doesn't mean you don't care.
If art is visually polished in some way such that quick quality cues are handled that doesn't mean that it will actually have a meaning or a message. You won't know until you have a chance to fully consume it.
The AI product doesn't fall apart (at least, not a good AI product.) The proposition is that is functionally indistinguishable from a human product to the extent that _there is no way to tell_, absent a formal and seperate disclosure.
It's a Turing test for art -- and the machines are winning, at scale.
> The fact that these same people can't tell whether it was authored by a human or an AI without a seperate disclosure is strong evidence that human authorship as such is not what they value at all.
By that logic, the fact that (some) people can't tell between gold and fool's gold is strong evidence that gold was not what they value at all.
Or: the fact that (some) people can't tell between food that is marketed to them as "healthy" and food that actually makes them healthy, is strong evidence that healthy food was not what they value at all.
> By that logic, the fact that (some) people can't tell between gold and fool's gold is strong evidence that gold was not what they value at all.
Well, no? Most people don't value gold for intrinsic attributes of the metal, but for the market-assigned value of it or perceived status.
I think GP overstated the point, but it's clear that a lot of people who object to AI authored content are objecting not because of the intrinsic attributes of the content but because of their emotions on the subject.
That doesn't make it invalid. If someone really doesn't want to consume AI generated content, then so be it.
But I also think the people making a big deal out of this seriously overestimate how deeply most people care. I suspect most people care only to the extent that the quality is lacking.
I'd justify that view with your example: People keep eating unhealthy food despite knowing it's bad. And as another example, we keep buying things we can't be sure is ethically sourced. We're willing to look past a whole lot.
The same way "Made in ..." labels were originally protectionist measures, I suspect disclosure requirements etc. will seriously backfire when it turns out most people stop caring the moment the quality is usually good enough.
I can tell you I hate more than anything when I’m lied to. Some AI art could be fantastic (haven’t seen anything yet but I’m sure it’ll happen someday).
I don’t want something you prompted an AI for because genuinely I could just do that myself. The barrier to entry is nil. I’m not interested in the idea an artist has I’m interested in how they execute it.
If they executed it through human effort then I place a much much higher value on that. Same as I would if someone selling a hand forged a knife rather than a mass produced cut out.
Even if the human-made item is lower quality I’d prefer it by a mile if for no other reason than I know someone’s love and effort went into it.
I could never have that appreciation for something a machine generated.
The lines might blur if an artist uses an AI to manipulate their own art that they made.
Point is - the output quality fundamentally doesn’t matter. It’s principle, not the output quality, that guides most people in avoiding AI art.
If you saw two paintings and one was painted with the blood and guts of the artist, and another was a nice oil painting - would you value the former more regardless of aesthetic merit?
Being unable to tell a game is garbage without personally playing it is a problem with or without AI. Though AI is vastly increasing the amount of slop out there, for sure.
It's not the preferred political narrative of the model that I worry about. It's how brazen they are about altering their models to achieve it. It makes me wonder what else they're altering. I have trust issues with OpenAI and Anthropic as well, but with those companies, at least I know their motives are purely profit driven. I don't have that assurance with xAI.
I don't think the comment you are replying to is giving up leverage. It's simply pointing to the OP that what he seems to be upset is not the thing itself, which everyone does, but only that he knows about it or ignores others doing the same.
When I say cancer sucks I'm not ignoring that other things also suck. So when people start trying to put those words into the words of others, they effectively act like cancer fanboys, which is weird and all about them, with no part of originating or relating to the the people they question about stuff they supposedly "ignore".
All LLMs inherit bias from their training data, and xAI’s argument is that Grok is being steered to counter that bias rather than simply inherit it. You can disagree with whether they succeed, but the act of steering isn’t automatically suspicious when knowing that every major model is steered. The relevant question is whether the steering moves the model closer to truth and neutrality, or just replaces one bias with another. At least with Grok, some of that intent is unusually explicit. I’d rather have multiple competing approaches to steering than a monoculture where every model quietly optimizes for the same idea of acceptable answers.
I don’t think that is necessarily a bad preference if this was an actual dichotomy. Not all types of manipulation is equal, and when you at least try to hide it shows at least some respect for the user.
That said, I don‘t believe this dichotomy is real. Personally I don‘t use AI, political manipulation is however only a relatively tiny part of my reasoning for opting out.
> It's how brazen they are about altering their models to achieve it.
We know all the models insert shadow prompts to nudge the answers in preferred political directions. How much more "brazen" can you get than that? Nobody is giving you fat-free results that just apply the models to your prompts.
More like a dictator versus a Poltiburo. If the “entire population” was voting on the shadow prompts of the other models they would look very different. Considering the recent election results, they would look more like Grok.
The dictator has a proven track record of stupid opinions in multiple topics, mostly programming, which directly can be measured and understood by people here.
Meanwhile the politburo is mostly nerds, who come and interact in places like hackernews.
So its basically having a moron making wild choices or a technocracy.
> Considering the recent election results, they would look more like Grok.
Considering that the largest voter base was "didnt vote" and that the voters of the republican party measured lower in literacy, technical knowledge, higher education acquisitions and even studies on accurately describing reality. I am not entirely confident they would participate or move the shadow prompt in any meaningful direction.
> Considering that the largest voter base was "didnt vote"
It’s a fallacy to treat “didn’t vote” as “didn’t support the winner.” Non-voters are more pro-Trump than average: https://data.blueroseresearch.org/hubfs/2024%20Blue%20Rose%2.... See p. 6 (“There’s a turnout story this cycle – but a different one than we’re used to talking about. With the combination of
less-engaged and less-likely voters leaning more GOP, a larger electorate meant a more Republican electorate. Projecting onto the full voter file, if every registered voter voted, it’s likely that Trump would have won by even more.”).
The data consistently shows that non-voters have lower trust in institutions. They’re the exact type of people who are going to be more skeptical of shadow prompt engineering being done by “safety experts” at Google and Meta.
I mean one of the guys got fired by his own board for lying and is still calling the shots. Another guy sued the Pentagon during a war and we're still letting him act like a nation state.
Musk's empire of personality cult is like, idk, on slightly more cocaine?
I'm having a hard time being like: "oh, that's the bad self-appointed, self-dealing would be God Emperor. they're not all like that. why some of my very best friends are cluster B psycho con men with crime funding."
How is changing history and making a a group of objectively white people black "removing bias" ˋ? What it is is literally bias. Like thinking Pi could be 4. Removing bias ends with truth, not these crazy wonky results.
If you're aruging about historical accuracy, but still want accurate looking generated images, I don't know what to say.
But to the technical point, A large part of the training corpus has biases that if left unchecked would cause PR based disasters for the company hosting it. ie the classic black teenager/white teenager.
Now as training of models is not an exact science, and neither is the fine tuning, its analogous to forcing a water balloon into a square box. Its possible but it has odd side effects when you get to the corners.
When making a _product_ you need to choose the least worse failure case. For grok it was for a long time, pandering to the ego of the owner. For Google, who is an advertising company, its about trying not to scare advertisers. This means everthing must be vanilla
So you have a huge number of photos of white people in the training data set, but other ethnicities exist. So to make the otherwise white-biased dataset less biased, you try to e.g. add a hidden system prompt that whenever the user asks for a group of people (unspecified ethnicity), it may instead ask for "mixed ethnicities" or whatever.
Ask for a group of Nazis, and that's it - this is how models work. No "LGBTQ liberal" propaganda is needed to explain it. Unlike what Musk is doing.
The training data and the end result should reflect the real world. No you don't need to "add bias to fight bias", just keep it truthful. Don't force SF-brained forced diversity/inclusion/whatever, we know well some models do.
> OpenAI invented a technique in July 2022 whereby its system would insert terms reflecting diversity (like “Black,” “female,” or “Asian”) into image-generation prompts in a way that was hidden from the user.
> Google’s Gemini system seems to do something similar, taking a user’s image-generation prompt (the instruction, such as “make a painting of the founding fathers”) and inserting terms for racial and gender diversity, such as “South Asian” or “non-binary” into the prompt
It seems then that their objective is to superficially increase the diversity of results, to avoid bad PR, rather than actually neutralising harmful and untruthful biases of the input data.
A model where asking for a math professor results in an image of an asian man, asking for an engineer in an image of a white man, and asking for a criminal results in an image of a black man
If you try to remove that in the name of "diversity" or being "less bigoted" you quickly end up with racially diverse nazis
That's the claim, and it's a belief that's self-fulfilling prophecy, like saying all politicians are corrupt.
If you can convince everyone that everyone is corrupt, it hurts anyone who isn't corrupt. You hear people preferring those who have no shame about their corruption, based on the premise that those who aren't overtly corrupt must be more sinister and dangerous if they hide their corruption so well.
Such a good point. Disinformation and trying to destroy sources of truth is just part of the puzzle. Often the worst damage is just "well, both sides are bad" -- because it just tells people not to listen to either side, or it is too much work.
The end goal: they want to push that you must sacrifice your rights to a monarch or authoritarian person for order and safety.
It's interesting that all models seems to be unbiased out of the box so far- that is, mostly reflecting the training data (the internet).
The whole mecha-hitler thing doesn't seem to reflect fine-tuning, it was just a prompt change.
There's been some studies that suggest that certain usage of LLMs reduces political bias, which seems reasonable. Like, how credible is climate change, are Haitians eating pets, etc. THings that have a basis in fact.
I don't put it past Elon to train a model with political bias, just that it hasn't happened yet.
> It's interesting that all models seems to be unbiased out of the box so far
This is really begging the question. If something relies on the perception of a human, it has bias. The data (or lack thereof) used to train models is per se a bias.
The mistake is assuming bias-removal is some virtuous goal to be achieved. It can't, and shouldn't. Alignment, while equally impossible, is at least a goal worth aiming towards.
I think the issue here is that there's no objective difference between "alignment" and "actively trying to shape its replies to fit a political narrative", beside the fact that the latter refers to the kind of political narratives that you don't like.
Sure, but there’s the direction of a vector, and the length of the vector.
xAI’s direction is hellish, and length is 100x any other provider’s. So, yeah, nobody is pure. But most are at least trying to be balanced and not just, you know.
That nazi salute how do you actually know it was ment to show off as a Nazi supporter? I mean who is telling this? Or is that your assumption based on something?
There is also his namesake, his racist father and his grandfathers past. There is a lot of background information available if you really look for it. And finally, if you look at story of "The Elon" and what Elon has been building, you could almost think that he is trying to actually live up to his namesake!
Trump's ex wife mentioned the only book he ever had in his bedside table at night was on hitler speeches. Multiple Trump aides have been caught reading the mEIN kampf. Stephen Miller is somehow the world's only nazi jew and writes trump's speeches.
To this cohort of people Elon spent a fortune on funding their campaign, even willing to commit election fraud (the 1 million giveaway case which is on going but seems open and close).
In front of that audience he did 2 nazi salutes chest to straight arm. He didnt apologise or explain it either.
What other possible explanation is there beyond "the dude saw a nazi adjacent politcal platform and spent hundreds of millions to make it succeed and then went mask off the second he knew there would be no repercussions"
Nah. My beliefs are actual truth, so if provider is shaping their models according to my preferred political narratives that's correct and only moral thing to do. Anything else would be morally bankrupt.
Who or what is this latter "globalist" view? Like it's such an oversimplification of multiple, independent actors' acts that it's just a plain bad world view/model. What would be the common goal? You need some extra amount of tinfoil for such bullshit.
No, but technically, all models go through an alignment phase where you feed data that's aligned to your goals and values to train the model so that it will exhibit the kind of behavior you want.
There's no "politically neutral" value system anyway.
Doesn't mean we shouldn't try to make the models more inclusive, less biased and less prone to extremism, but in a technical sense yes, actually everyone does it.
And it's all the same to you? You don't care which values those things have? Of course there will always be underlying values. I wouldn't go as far as calling everything political narratives.
I think it's just pretty clear that Elon's values are not what most people want the world to be shaped by.
That's reductive, how many other models had a mecha hitler incident? Or a "let's talk about white genocide in south africa" incident?
There is some truth to what you say, but most model providers I would say are engaged in CYA type shaping moreso than anything, grok is actively and openly being developed to spread a white nationalist agenda. There are levels to this.
False, most are just throwing in data into the pipeline and hoping something very smart comes out the other end. Their shaping is to have it code very well, beat all the benchmarks and be safe enough that the powers that be don't and public sentiment doesn't turn against them.
Sure, every author has a bias. But a fair selection of human written sources will be pretty balanced (of course, given the ratio of languages, surely western models will have a western-christian bias - presumably Chinese models less so, but this latter I have no way of checking).
If you train specific stuff on top, or deliberately filter the sources (e.g. Tiananmen square), your model is deliberately less honest on that topic. Grok is probably the worst in deliberately filtering and training "out" specific stuff (to the point where Elon posted stuff that "they will 'fix' the models" real output when it said something true but remotely liberal). Claude and chatgpt definitely have some similar stuff, but mostly to protect themselves (e.g. suicide prevention, not saying slurs, etc). I don't think the two is comparable (reality bending vs basic etiquette-kind of not saying everything out loud)
Yes. But only one trillionaire is screaming from twitter that he will use his wealth to combat the woke mind virus by tuning his model. So, ya know, there’s THAT.
Yeah but their political narratives involving dehumanizing people, and hurting minorities - which is way worse. The attempt at false equivalence between traditional American propaganda and right wing American propaganda is disgusting.
Maybe, but some political narratives are good and right and true, and others are bad and wrong and false. (I am not joking: objective reality exists and most of politics isn't subjective.)
It has occurred to the OP, but those ones are ok as they fit OP's political leanings. They just want to silence other political narratives that dont agree with theirs, they want Grok in the Gulag.
No. Other companies are altering thier products to chase market share and profits. Grok is run by someone beyond profit motives who is actively promoting a personal agenda. I would rather work with Microsoft's counsel of dark wizards than an individual with an axe to grind with the world.
From a business perspective, a company that trains it's LLMs to having boring, mainstream, generally-inoffensive views is a big selling point over whatever the hell Elon is doing.
Drugs. Drugs are what Elon is doing. And it's pretty cringe.
But it's remarkably similar in cringe to that little "secret erection" look Amodei gets when he talks about millions of unemployed people, or or Altman rolling through Pacific Heights in a four million dollar Swedish hypercar holding the steering wheel wrong the day after yet another lecture about UBI.
You only get to become one if the idea of spending 60 hours a week typing code into a computer during the late 90's/early 2000's is palatable to you. This causes a dweeb selection funnel.
Despite the whirlwind of media to the contrary, the US is very welcoming to foreigners who follow the laws (that is, don't enter illegally) and make an effort to integrate by learning the language and customs.
Immigrating to Canada was a breeze for decades until the last 2-3 years when they started to reduce immigration numbers in response to citizen concerns.
Eh it's not as black and white as you make it look. Colleague of mine is in ICE detention, because ICE acted on courts being held up in appeals so they can ignore his ongoing asylum case and deport him back to Russia. He followed the rules, had a work permit and everything, but did not matter in the end.
The green card interview snatching is also messed up, if your existing visa expired, while you were being processed, USCIS was understanding and it did not affect your application. (Processing time is slow, so that can happen). Now it's if your visa ever expires your Freiwild for ICE. They're technically not wrong with this, but they're essentially throwing the book at people getting a visum legally too.
That doesn't contradict what I'm saying. They aren't going to do anything that might depress the economy—it's just a sufficient cruelty to satisfy their base.
Leetcode merely tests your ability to memorize and regurgitate algorithms from a textbook -- a skill that LLMs can now do a thousand times faster and more accurately than any human.
It never had much to do with real world software engineering, and now it's even less relevant than ever.
reply