site banner

Culture War Roundup for the week of October 3, 2022

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

24
Jump in the discussion.

No email address required.

I think over the last few months we've established that AI issues are on topic for the culture war thread, at least when they intersect with explicitly cultural domains like art. So I hope it's ok that I write this here. Feel free to delete if not.

NovelAI's anime model was released today, and it's pretty god damned impressive. If you haven't seen what it can do yet, feel free to check out the /hdg/ threads on /h/ for some NSFW examples.

Not everyone is happy though; AI art has attracted the attention of at least one member of congress, among several other public and private entities:

WASHINGTON, D.C. – Today, U.S. Rep. Anna G. Eshoo (D-CA) urged the National Security Advisor (NSA) and the Office of Science and Technology Policy (OSTP) to address the release of unsafe AI models that do not moderate content made on their platforms, specifically the Stable Diffusion model released by Stability AI on August 22, 2022. Stable Diffusion allows users to generate unfiltered imagery of a violent or sexual nature, depicting real people. It has already been used to create photos of violently beaten Asian women and pornography depicting real people.

I don't really bet on there being any serious legal liability for Stability.AI or anyone else, but, you never know.

I've tried several times to articulate here why I find AI art to be so upsetting. I get the feeling that many people here haven't been very receptive to my views. Partially that's my fault for being a bad rhetorician, but partially I think it's because I'm arguing from the standpoint of a certain set of terminal values which are not widely shared. I'd like to try laying out my case one more time, using some hopefully more down-to-earth considerations which will be easier to appreciate. If you already disagree with me, I certainly don't expect you to be moved by my views - I just hope that you'll find them to be coherent, that it seems like the sort of thing that a reasonable person could believe.

Essentially the crux of the matter is, to borrow a phrase from crypto, "proof of work". There are many activities and products that are valuable, partially or in whole, due to the amount of time and effort that goes into them. I don't think it's hard to generate examples. Consider weight lifting competitions - certainly there's nothing useful about repeatedly lifting a pile of metal bricks, nor does the activity itself have any real aesthetic or social value. The value that participants and spectators derive from the activity is purely a function of the amount of human effort and exertion that goes into the activity. Having a machine lift the weights instead would be quite beside the point, and it would impress no one.

For me personally, AI art has brought into sharp relief just how much I value the effort and exertion that goes into the production of art. Works of art are rather convenient (and beautiful) proof of work tokens. First someone had to learn how to draw, and then they had to take time out of their day and say, I'm going to draw this thing in particular, I'm going to dedicate my finite time and energy to this activity and this particular subject matter rather than anything else. I like that. I like when people dedicate themselves to something, even at significant personal cost. I like having my environment filled with little monuments to struggle and self-sacrifice, just like how people enjoy the fact that someone out there has climbed Mt. Everest, even though it serves no real purpose. Every work of art is like a miniature Mt. Everest.

Or at least it was. AI art changes the equation in a way that's impossible to ignore - it affects my perception of all works of art because now I am much less certain of the provenance of each work*. There is now a fast and convenient way of cheating the proof of work system. I look at a lot of anime art - a lot of it is admittedly very derivative and repetitive, and it tends to all blend together after a while. But in the pre-AI era, I could at least find value in each individual illustration in the fact that it represented the concrete results of someone's time and effort. There are of course edge cases - we have always had tracing, photobashing, and other ways of "cheating". But you could still assume that the average illustration you saw was the result of a concrete investment of time and effort. Now that is no longer the case. Any illustration I see could just as easily be one from the infinite sea of AI art - why should I spend any time looking at it, pondering it, wondering about the story behind it? I am now very uncertain as to whether it has any value at all.

It's a bit like discovering that every video game speedrun video you see has a 50% chance of being a deepfake. Would you be as likely to watch speedrunning videos? I wouldn't. They only have value if they're the result of an actual investment of time by a human player - otherwise, they're worthless. Or, to take another very timely example, the Carlsen-Niemann cheating scandal currently rocking the world of chess. Chess is an illustrative example to look at, because it's a domain where everyone is acutely aware of the dangers of a situation where you can't tell the difference between an unaided human and a human using AI assistance. Many people have remarked that chess is "dead" if they can't find a way to implement effective anti-cheating measures that will prevent people from consulting engines during a game. People want to see two humans play against each other, not two computers.

To be clear, I'm not saying that the effort that went into a work of art is the only thing that matters. I also place great value on the intrinsic and perceptual properties of a work of art. I see myself as having a holistic view where I value both the intrinsic properties of the work, and the extrinsic, context-dependent properties related to the work's provenance, production, intention, etc.

TL;DR - I used to be able to look at every work of art and go "damn someone made that, that's really cool", now I can't do that, which makes every interaction I have with art that much worse, and by extension it makes my life worse.

*(I'm speaking for convenience here as if AI had already supplanted human artists. As I write this post, it still has limitations, and there are still many illustrations that are unmistakably of human origin. But frankly, given how fast the new image models are advancing, I don't know how much longer that will be the case.)

EDIT: Unfortunately, this dropped the day after I wrote my post, so I didn't get a chance to comment on it originally. Based on continually accumulating evidence, I may have to retract my original prediction that opposition to AI art was going to be a more right-coded position. Perhaps there are not as many aesthetes in the dissident right as I thought.

EDIT: Unfortunately, this dropped the day after I wrote my post, so I didn't get a chance to comment on it originally. Based on continually accumulating evidence, I may have to retract my original prediction that opposition to AI art was going to be a more right-coded position. Perhaps there are not as many aesthetes in the dissident right as I thought.

I suspect that pro-/anti-AI art won't cleave neatly along existing ideological lines, but I do think for now, there's good reason to believe that pro-/anti- will be right/left respectively. The 2 dominant factors that would drive this are (1) most pre-existing artists are on the left and (2) AI art makes censorship tougher. Why (1) would lead those on the left to be against AI art has been expounded upon plenty in this very thread. For (2), when it comes to the most mainstream instantiations of art, i.e. pop culture, the dominant narrative for the past decade+ from the left has been that art that doesn't fit neatly within certain boundaries can cause literal harm to real humans and thus must be censored or at least censured in some meaningful way, for the prevention of harm to innocents. AI art - or more specifically easy access to AI art tools - would open up the floodgates to everyone being able to create art that is deemed harmful, in a way that our current censorship technology just couldn't keep up with.

I wonder if this will hold in the long run. There are plenty of leftist reasons to support AI art (e.g. more access to sophisticated art creation by people who otherwise would be unable to do so) and rightist reasons to be against it (e.g. traditional and/or religious reasons of being against degeneracy which AI art would enable like nothing before). It's just basically impossible to predict the way things will go when it comes to which subgroups within each tribe will win out.

[...] on topic for the culture war thread [...] I hope it's ok that I write this here. Feel free to delete if not.

As far as I can tell, anything is on-topic for the culture war thread if it's interesting and well written.

It doesn't seem all that much more effortless than many of Robert Ryman's works.

I'm sure someone will find a way to get their vision in AI assisted art creation featured in a gallery or sold for a princely sum.

Do you make art?

I make art as a hobby and teach it, and feel moderately positive toward the recent developments in AI art.

There are a couple of different things that will become more obviously different. There's commercial art, which will likely be extensively created by AI in the fairly near future. The automation of anime nudes hardly seems like a loss worth mourning. There's high status Artist art, which will not change all that much, and already isn't much about visual skill, so much as social skill. There's popular art, which might become some kind of combination thing, with different classifications and disclaimers. There's gift art, which is almost entirely about effort and thoughtfulness, and not much about skill. This seems intrinsic in children as soon as they can talk, and won't be changing much.

Personally, I like the process of art making more than artistic artifacts, and am generally uninterested in artwork that clearly took painstaking detail oriented labor. There are photorealists who show off by making 100 hr paintings of extremely detailed faces or whatever, and I understand caring about that, but do not care about it myself. This seems unlikely to be faked very often -- process videos are already very popular, and will likely become even more so. There is not enough status at stake, and it's rather niche. There isn't really any reason you couldn't still find detailed realistic artists practicing their craft.

There's a quote attributed to Picasso that "when art critics get together they talk about content, style, trend and meaning, but when painters get together they talk about where can you get the best turpentine." I like paint and wool and cold pressed cotton paper and warm wax and translucency and the smell of certain mediums and the changes that pottery undergoes as it progresses through multiple firings. I'm excited that there are now water mixable oil paints (no turpentine required!) and Derwent ink pencils. These artisanal practices have already been stripped of most of their importance. They are crafts, practiced by retired ladies in their craft sheds. They are unserious. Plenty of visual art is already like that as well. Hobbyist empty nesters painting impressionist oils of the local wildlife. This is a bit dispiriting, but will not be meaningfully changed by AI. Children will still always give something they made to their family, old ladies will still paint Monet knock offs of their regional landscape. These phenomena are not primarily about the image as such anyway, but about the process and physical manifestation of love or attention.

I really don't find your post convincing in the least. And the constant whiny bitching and crying by artists about AI art has made me suspicious of the motivations of these so called "artists", who claim to do it so much so for their love of the art. @EfficientSyllabus, said it a lot better than me. You are not lamenting the loss of an artform, you are lamenting the loss of status.

Here are a few scattered as to why I am so deeply unsympathetic to those who endlessly moan about AI art.

  • I'm a programmer. I believe I love the art of programming. I also know some people who genuinely love the art of programming. When I saw what OpenAI Codex (AI that can generate code) was capable of doing, My jaw was on the floor. A program that could write more programs?? It was science fiction in-front of my own eyes. Every other programmer I know who loves the game itself had the exact same reaction; amazement.

    So what if a machine can write code? Code is good! The world needs more code! Code makes machines more efficient, it does boring jobs that people would have to do, code optimizes processes that literally puts food into most peoples mouths.

    In the same vein? Is art not a good thing? Is the world not a richer place because there will be more art? Isn't it great that an independent blogger who couldn't afford commissions will now get to have art that makes his blogging richer? Is it not great that a mom and pop shop can now produce artwork that will make their corner store more lively? Won't the world get a little bit more aesthetically pleasing?

    Why are the majority programmers so enthusiastic about machines that can code but not artists?

    Maybe because the greatest trick the devil pulled was that "artists" are in it for the love of the art and us uncool dirty nerds are in it for the money and status?

  • There is an art to almost every process right?

    Farming can also be an art right? Getting the soiled tilled just right, making sure the seeds are placed just the appropriate distance apart, etc.

    However, if someone lamented the loss of farming as an artform because combine harvesters were invented... My and hopefully any rational persons response would be;

    " You motherfucker. Do you not realize that millions of hungry mouths will be fed because of this thing? Is your artsy fartsy shit more important that people not being hungry?"

  • The world is a place where things need to get done.

    I love the art of programming and spend countless hours cleaning up my programs, but ultimately it's of no value if no one can use my programs. Chefs can put their heart and soul into their food, but it would be of no value if no one ate it.

    The value is in the PRODUCT, not the PROCESS.

    If my favorite bakery found a way to mass produce their cheesecakes but the pastry chef was not required anymore and it would be all done by machines, then good. More people can enjoy great food for cheap. And to be honest my tongue doesn't care, if it did, its priorities are not in order.

    Boohoo for the pastry chef, if they love making cakes so much they can make the cake and throw it in the trash. In my world cakes are for eating. Is it not wisdom that you cook for your friends and family for them, not for you? The sanctity is in the fact that their stomachs are full not that your knife skills are perfected?

    Same for the artists, they can draw their art and throw it in the trash, its the process that matters right?

As someone who believes that more things are good. Products are good. Anyone lamenting about a process that brings more good things into the world is my enemy. You are actively lamenting that the world is becoming a richer place, in both the very economic and metaphorical sense of the word 'richer'. The pie is getting bigger, you are just lamenting you won't be having a relatively larger share of it.

I think there is a certain line of thinking that can plausibly be raised as a defense:

  • The process of freely creating art is valuable as a form of human expression, either per se or because it enriches the human experience in some way. (This position is apparently one which you do not hold, but let's assume for the moment that a large portion of the population does sincerely hold it.)

  • So far, the process of creating art has been subsidized by its products which can be sold: corporate art, commissions, etc. However, in the future, these products are poised to be far more efficiently generated by AI.

  • Without revenue from these products, many of today's artists will be forced to move into other fields, and perhaps curtail their personal output due to no longer having enough time, supplies, or practice. This is bad, since it decreases the quality and quantity of valuable art creation.

  • Similarly, once the creation of art becomes no longer profitable, the second-order effects start to occur: the entire industry of art education gradually falls apart, and many people become unable to learn the skills to express themselves through art in the way they would prefer.

That is, the process of art-as-human-expression will be impacted negatively by the AI-driven devaluing of art-as-a-commercial-product.

I recall a discussion on LW or SSC (that I am now unable to find), about how many try to find economic justifications for avoiding animal stress, looking for evidence that less-stressed cows (for instance) produce better meat, since that kind of justification is the only form our society will accept: if no such justification can be found, then animal welfare will inevitably get tossed out the window. I interpret @Primaprimaprima's perspective in a similar light; if there is no more value in humans creating art as a product, then there will be nothing left to prop up the tradition of art as an expression, and the world will be worse off for it.

(Whether this assumption of art-as-expression depending on the existence of art-as-a-product holds up in reality is a different question. But it certainly seems like a plausible enough risk to worry about, assuming one values art-as-expression.)

If art as expression is valuable enough then people will spend money to do it for its own sake. If it is not then they were being subsidized the whole time and it was less valuable than we previously thought. I think art will survive but most artists will need to get dayjobs, just like all the people who used to subsidize their art who didn't personally get to experience the expression element. Because that's really the hidden cost that isn't being brought up, the tiny tax on everyone else in society so that artists could be for lack of a better word "unproductive"(undoubtedly they were productive in a pre-ai-art world but they are no longer productive in a post-ai-art world)

Indeed. I suppose that the next step of the defense would be that society persistently undervalues art-as-expression: if the general public were aware of its full value, they would pay for art-as-expression, but structural factors and lack of quantifiable benefits makes awareness implausible in the near future. (Compare this to the animal-welfare activist who fights against factory farmers' greed and consumers' apathy: they believe that if the public were aware of the full value of animal welfare, then animal-protection laws would be passed in a heartbeat.)

In this scenario, the best outcome, short of formal subsidies for artists, would perhaps be a large-scale donation model, much like for many orchestras and museums today. But this is still much less accessible to artists than the pre-AI status quo, where art-as-expression maintains a safe existence as a byproduct of art-as-a-product. So it would still make sense for those who value art-as-expression to lament this change beyond the effects on their own lifestyles, given that this particular Pandora's Box isn't getting closed any time soon.

I think there are two distinct divisions here that are worried or not worried about AI art.

(1) The 'fine artists', producing the likes of Basquiat painting in that story further down about the Guggenheim. The very top ones won't be affected, hell they may even get into using AI to produce art, because it's all about the concept and not the actual work. Damien Hirst did not cast himself works like this, he does the design then hands it off to a foundry to make it. So AI art is not a threat to art which is about concept, notoriety, ethnicity, who is the latest hot property taken up by the galleries and rich collectors, etc.

(2) The commercial artists, who very well may find themselves out of a job if AI can churn out made-to-measure works for posters (like that awful German Green Party one), magazine and online article illustrations in this style called Alegria or Corporate Memphis, advertising and product art, and the rest of it. Some of them can adopt it as one more tool, like the software they already use, but if a big corporation can create its own in-house art by purchasing an AI program to do it, then that cuts out freelancers and those who rely on commissions. Amateur artists are a sub-set of this, all the artists doing fan-art for commission may be priced out if you can instead get to use an AI who will do exactly what you want the way you want it.

So there is definitely a panic about "the AI is taking our jobs!" and that isn't completely mockery, because there will be people who can no longer make a living doing commercial art. How that shakes out remains to be seen, and we really won't know until AI art is widely used. Maybe people will go back to having a Real Human Drew This piece of work, to stand out from all the mass-produced AI art, especially for things like fashion magazines that want to sell themselves as being creative and different and unique.

And there are real concerns about art as art, from people who enjoy creating art and don't like the implication that this is just one more human activity that can be mechanised and turned into extruded product. You say that the value is in the product, not the process, but for most of us our experience of mass-market mechanised production of, for instance, food products has not been "oh wow, this cheesecake is so delicious and gorgeous, just like a pastry chef made it!", it has been "replace ingredients with cheapest substitute, lots of artificial flavouring and colouring, and a process that is economically convenient for the manufacturer" ending up in bland, processed, 'not as good as the real thing' goods (see the furore over how Cadbury chocolate has changed since Mondelez bought it).

If the experience of mechanisation was "wow, gorgeous!" instead of "yeah, now it's gonna be cheap, bad-tasting gunk", then people would be less alarmed about AI art (as distinct from the financial element). You say "More people can enjoy great food for cheap. And to be honest my tongue doesn't care, if it did, its priorities are not in order" but would you really not care if it tasted different? Why is that bakery your favourite bakery, if not for the very reason that it pleases your tongue? "Okay, now the cheesecake tastes like chalk and mouse-droppings and gives me diarrhoea after I eat it, but shut up tongue! The process is more efficient and cheaper and productive, who cares about the quality of the end product?"

If the experience of mechanisation was "wow, gorgeous!" instead of "yeah, now it's gonna be cheap, bad-tasting gunk", then people would be less alarmed about AI art (as distinct from the financial element). You say "More people can enjoy great food for cheap. And to be honest my tongue doesn't care, if it did, its priorities are not in order" but would you really not care if it tasted different? Why is that bakery your favourite bakery, if not for the very reason that it pleases your tongue? "Okay, now the cheesecake tastes like chalk and mouse-droppings and gives me diarrhoea after I eat it, but shut up tongue! The process is more efficient and cheaper and productive, who cares about the quality of the end product?"

So artists are losing people without taste who were being overcharged for what they were experiencing, people with taste will stick to real artists. What's the problem here? I can still get and do still get fancy hand crafted artisanal food, it costs as much as it always has. But now the poor can get at least an approximation for cheap enough for them to afford. I imagine high profile publications and AAA game titles will still have humans doing their art for them, but suddenly indie publications and indie games can afford as many art assets as they can productively use. This is a pure win for expression.

And the constant whiny bitching and crying by artists about AI art has made me suspicious of the motivations of these so called "artists"

Write like you want to include everyone in the conversation, please. This is unnecessarily heated.

But, critically, that isn't the actual argument being made by artists, probably because it's a losing argument. Milton Friedman's classic story:

While traveling by car during one of his many overseas travels, Professor Milton Friedman spotted scores of road builders moving earth with shovels instead of modern machinery. When he asked why powerful equipment wasn’t used instead of so many laborers, his host told him it was to keep employment high in the construction industry. If they used tractors or modern road building equipment, fewer people would have jobs was his host’s logic.

“Then instead of shovels, why don’t you give them spoons and create even more jobs?” Friedman inquired.

This is probably the key point. It’s impossible for a non-programmer to do anything useful with Copilot. Non-artists can already do useful things with NovelAI right now though; they’re ready to start cutting artists out of the loop now, today. It’s not a comparable situation.

Why are the majority programmers so enthusiastic about machines that can code but not artists?

Because they aren't. They're collectively deluding themselves into believing in the «soul» and that programming will never be automated by AI. Just like certain artists are.

I am a programmer. OpenAI scares me. I'm putting every effort I've got into the Grind, because I think the industry's due for a phenomenal crash that'll leave the majority in the dumps. You are free to disagree.

There is no problem humans face that cannot be reframed as a programming or automation problem. Need food? Build a robot to grow it for you, and another to deliver it to your house. Need to build a robot? Make a factory that automates robot fabrication. Need to solve X medical issue? Write a program that figures out using simulations or whatever how to synthesize a chemical or machine that fixes it. Given this, the question of "what happens to programmers when computers can write code for arbitrary domains just as well as programmers can" answers itself.

I expect that fully automating coding will be the last job anybody ever does, either because we're all dead or we have realized Fully Automated Luxury Space Communism.

Is there something misleading with the way I phrased my comment? I don't understand why multiple people have succeeded in reading "programmers will be completely replaced by AI" into my words.

And this isn't a nitpicking thing. It is an extremely important distinction; I see this in the same way as the Pareto Principle. The AI labs are going to quickly churn out models good enough to cover 95% of the work the average software engineer does, and the programming community will reach a depressive state where everyone's viciously competing for that last 5% until true AGI arrives.

Your first paragraph misses how hard it is for human programmers to achieve those things, if it is even possible under current circumstances (find me a program that can acquire farmland & construct robots for it & harvest everything & prepare meals from raw materials). Even hiring an army of programmers (AI or no) would not satisfy the preconditions necessary for getting your own food supply, namely having an actual physical presence. You need to step beyond distributed human-level abilities into superhuman AI turf for that to happen.

There is a sense in which the job of coding has already been automated away several times. For instance, high-level languages enable a single programmer to accomplish work that would be out of the grasp of even a dozen assembly-language programmers. (This did, in fact, trash the job market for assembly-language programmers.)

The reason this hasn't resulted in an actual decline in programmer jobs over time is because each time a major tool is invented that makes programming easier (or eliminates the necessity for it in particular domains), people immediately set their sights on more-difficult tasks that were considered impractical or impossible in the previous paradigm.

I don't really see the mechanism by which AI-assisted programming is different in this way. Sure, it means a subset of programming problems will no longer be done by humans. That just means humans will be freed to work on programming and engineering problems that AI can't do, or at least can't do yet; and they'll have the assistance of the AI programmers that automated away their previous jobs.

And if there are no more engineering or programming problems like that, then you now have Automated Luxury Space Communism.

Roughly speaking, I see your point and agree that it's possible we're just climbing a step further up on an infinite ladder of "things to do with computers".

But I disagree that it's the most likely outcome, because:

  1. I think the continued expansion of the domain space for individual programmers can be partially attributed to Moore's Law. More Is Different; a JavaScript equivalent could've easily been developed in the 80s but simply wasn't because there wasn't enough computational slack at the time for a sandboxed garbage collected asyncronous scripting language to run complex enterprise graphical applications. Without the regular growth in computational power, I expect innovations to slow.

  2. Cognitive limits. Say a full stack developer gets to finish their work in 10% of the time. Okay, now what? Are they going to spin up a completely different project? Make a fuzzer, a GAN, an SAT solver, all for fun? The future ability of AI tools to spin up entire codebases on demand does not help in the human learning process of figuring out what actually needs to be done. And if someone makes a language model to fix that problem, then domain knowledge becomes irrelevant and everyone (and thus no one) becomes a programmer.

  3. I think, regardless of AI, that the industry is oversaturated and due for mass layoffs. There are currently weak trends pointing in this direction, but I wouldn't blame anyone for continuing to bet on its growth.

For (1), what you're saying is certainly true; the better abstractions and better tooling has been accompanied by growth in hardware fundamentals that cannot be reasonably expected to continue.

(2) is where I'm a lot more skeptical. A sufficient-- though certainly not necessary-- condition for a valuable software project is identifying a thing that requires human labor that a computer could, potentially, be doing instead.

The reason I called out robotics specifically is because, yeah, if you think about "software" as just meaning "stuff that runs on a desktop computer", well, there's lots of spheres of human activity that occur away from a computer. But the field of robotics represents the set of things that computers can be made to do in the real world.

That being so, if non-robotics software becomes trivial to write I expect we are in one of four possible worlds:

World one: General-purpose robotics-- for example, building robots that plant and harvest crops-- is possible for (AI-assisted) human programmers to do, but it's intrinsically really hard even with AI support, so human programmers/engineers still have to be employed to do it. This seems like a plausible world that we could exist in, and seems basically similar to our current world except that the programmer-gold-rush is in robotics instead of web apps.

World two: General-purpose robotics is really easy for non-programmers if you just make an AI do the robot programming. That means "programming" stops being especially lucrative as a profession, since programming has been automated away. It also means that every other job has been (or will very soon be) automated away. This is Fully-Automated Luxury Space Communism world, and also seems broadly plausible.

World three: General-purpose robotics is impossible at human or AI levels of cognition, but non-robotics AI-assisted programming is otherwise trivial. I acknowledge this is a world where mass layoffs of programmers would occur and that this would be a problem for us. I also do not think this is a very likely scenario; general-purpose robotics is very hard but I have no specific reason to believe it's impossible, especially if AI software development has advanced to the point where almost all other programming is trivial.

World four: World two, except somebody screwed up the programming on one of their robot-programming AIs such that it murders everyone instead of performing useful labor. This strikes me as another plausible outcome.

Are there possibilities I'm missing that seem to you reasonably likely?

For your point (3), I have no particular expectations or insight one way or another.

Hi, I just want to leave a stub response: you seem right and I failed to type a recent response after reading 2 days ago.

We've been trying to innovate ourselves out of a job since the very beginning. I work with high powered business people that frequently can't even manage the most basic computer tasks, let alone automate them. We'll have a niche so long as people continue to work. What a glorious day it will be when all that intelligence and ingenuity is put to tasks other than making ads serve 0.2% faster.

You really think AI is going to replace programmers? If it does then it will be smart enough to self-modify, and then career concerns are the least of our worries.

If it does then it will be smart enough to self-modify,

This does not work out the way you think it will. A p99-human tier parallelised unaligned coding AI will be able to do the work of any programmer, will be able to take down most online infrastructure by merit of security expertise, but won't be sufficient for a Skynet Uprising, because that AI still needs to solve for the "getting out of the digital box and building a robot army" part.

If the programming AI was a generalised intelligence, then of course we'd be all fucked immediately. But that's not how this works. What we have are massive language models that are pretty good at tackling any kind of request that involves text generation. Solve for forgetfulness in transformer models and you'll only need one dude to maintain that full stack app instead of 50.

What I'm saying is that AI's are made of code. If they can write code then they can improve themselves. An AI able to code better than people can also code a better AI than people can. Maybe you don't think that that will lead to recursive self-modification--I think there's at least a good chance that there are diminishing returns there--but just consider the advances we've made in AI in the last year, and you're supposing a future where not only have we gotten farther but then there's another entity capable of going farther still. At a bare minimum I think an AI capable of doing that is capable of replacing most other careers too.

I too am a programmer, but fortunately the German software industry is so far behind the times and slow to evolve and German employment laws in general are so strict in their regulations in favor of employees that I think I can safely coast halfway to retirement before I feel any market pressure.

That depends on how much of a difference AI will make, doesn't it? If advanced AI enables big American corps to churn out absurdly efficient code or highly advanced machine designs in minimal time, what will sclerotic German companies do?

I used to work at Siemens and half the people employed as programmers there thought that automating things in Excel was black magic, let alone doing basic things in Python with libraries like pandas. The difference in productivity compared to its rivals is small enough that coasting on momentum of past strengths might be sufficient to stay relevant in the present, but strong AI could plausibly make a lot of crusty German institutions obsolete in a way that our lawmakers won't be able to compensate for.

Stop scaring me. If the statists are going to tax me anyways then the least I expect to receive in exchange is the illusion of security.

Why are the majority of programmers so enthusiastic about machines that can code

Because they have no foresight.

I write code for a living. Of course I don’t want machines to learn how to code. That would put me out of a job! Why would I want someone to build something that would put me out of a job? That makes no sense.

I assume that by the time machines have truly automated coding (that means they can debug and fix your 40 year old proprietary system too according to natural language requirements, not just generate new code) then we’ll have AGI. I’ve become even more pessimistic about AGI lately and I think it’s likely that it will lead us straight into a dystopia, because there’s no reason for the rich to give us UBI. Once they no longer need us, they’ll more than likely just let us starve. So I hope to God that machines don’t automate coding in my lifetime.

I find this pessimism ridiculous, what are the rich going to do with all that corn? Everything I've seen or heard about them is that they want to be admired and loved by the people, and I at least live in a democracy where it's not totally up to the rich to decide to let us all starve.

I'm of the same opinion. If the future is fully automated, then one doesn't need the AI itself to go bad in order to have a humanitarian catastrophe - it will be sufficient for those who command the AIs to determine that keeping the unproductive classes as pets is no longer fashionable.

This is my belief about the dangers of AI - it's not the AI itself (because I don't think we will get self-aware, having goals and wanting to meet them, agentic like a person AI) but the people who use it, control it, and think they can depend on it to solve problems like "best economic policy" or 'let's create the technocratic utopia'.

Let's say Boston Dynamics creates a genuinely humanoid robot that can do all the things humans do, and Amazon buys a ton of them to replace the human workers in its fulfilment centres (particularly since they have concerns about labour shortages ). Great, this is way more efficient, Amazon can now sell us "same hour delivery", costs go down and productivity goes way up, things are better all round for everyone!

Except the laid-off warehouse workers, because where are they going to find jobs? How are they going to live? Even 'learn to code' isn't enough anymore, because we have AI to do that, too. And Amazon is not going to pay them wages or the equivalent in UBI out of the goodness of its heart, why would it? It's a business, not a charity. The government may have to do this, and it will raise money by taxing Amazon, and Amazon will use the usual legal loopholes and run-arounds to avoid paying more tax than the minimum necessary (tax avoidance, not tax evasion is the distinction I have been told) so where will the money come from? If we think the pensions shortfall is going to be a massive shock, wait until you can't even work until you're 80 as a Walmart greeter to supplement your pension or whatever money you have to live on, because there's a BD robot doing that now,

tax avoidance

Tax avoidance works because we have literal laws on the books that let them do it, usually for quite good reasons. If the tax revenue is insufficient we can simply eliminate these ways to reduce tax burden or raise taxes.

If my favorite bakery found a way to mass produce their cheesecakes but the pastry chef was not required anymore and it would be all done by machines, then good. More people can enjoy great food for cheap. And to be honest my tongue doesn't care, if it did, its priorities are not in order.

I think you go too far here. If your tongue cares, it's because the machines are not producing the same cheesecakes you enjoyed from the pastry chef, they are producing almost the same. This is great for everyone who wants a good cheesecake for a reasonable price, but it is also good for the pastry chef - now he gets to make cheesecakes for people who want cheesecakes that are better than the machines can make. Plus he gets the fame and prestige of being the model upon which the machines are based. Nobody is putting "Made by machines!" on the label of their cheesecakes - they want a cheesecake made by the pastry chef for a reasonable price. Thanks to the machines they can get it, and if that's not good enough then they can pay for the chef's actual cheesecake.

They are equal in my analogy.

What a strange thing to say. Never mind that mass production has been around for a century and never ever created products identical to handmade, what other priorities might your tongue have then? Have you considered that maybe your tongue doesn't actually care, but due to a minor addendum you made to one of its arguments - helping it no less - it is just pretending to care because it is a child?

They are equal in my analogy

But in reality? That is the difference between people's experience of increasing automation and industrialisation, and the rosy forecasts of "by the 1980s, people will have so much leisure time it will be hard to fill it all, because the work week will be hours not days, thanks to machines!"

Unless we get Star Trek style replicators, the machine-made cheesecake will never be equal to 'the real thing' (and even in Trek, people still go out to restaurants where humans do the cooking). There's even an entire Youtube channel with different levels of chefs making different dishes - here's one for cheesecakes. This is why people pay different prices for different levels of cooking - you don't expect premium prices for fast food burgers, and you expect a higher level of quality if ordering a steak in a fancy restaurant.

I believe there is an American expression, used pejoratively, about "whitebread" or "Wonder bread", deriving from commercially produced sliced white bread loaves, filled with flour improvers and preservatives to enable it to remain soft and long-life. Now this is decried as spongy, tasteless and inferior. These were created thanks to the [Chorleywood Process}(https://en.wikipedia.org/wiki/Chorleywood_bread_process) which gave rise to the expression "the best thing since sliced bread" since the innovation was new and remarkable and consumers loved the product.

But I think anyone will agree that the commercial sliced panloaf is not as tasty as the bakery loaf, even if it lasts longer and is ready-sliced. I use both, I prefer batch bread. The constant 'improvements' turned out not to be improvements but were certainly an economically superior process. The end product suffered. That is the fear around AI art.

Wonderbread is a real brand and product that exists and is popular, not a pejorative. I prefer it for some types of sandwiches, like a pb&j and not for others. The world is richer for the existence of wonderbread.

Same for the artists, they can draw their art and throw it in the trash, its the process that matters right?

I think there have been instances of artists destroying their work or making art that essentially self-destructs (e.g. KLF, that one MMO art game where the game would shut down if people killed each other enough times or something, I think there was an installation piece that would beat itself apart), so really, they're kind of ahead of you on that.

My point was that if the process is what matters, there is nothing to be afraid of. No one cam steal that from you.

I too am a programmer, and am horrified by Copilot and friends. I write code to solve problems and release it under copyleft so that people can modify it for their own ends and share alike. I don't release it for it to be bundled up into some training set for a system that will accelerate the generation of non-free software.

Whatever an artist's goal in developing a skill, I think it's fair for him to be utterly crushed at the thought of his artistic career and personal style being reduced to an "by artist X" prompt to an image generator.

Sure if they were honest and just said "im not pleased about losijg my job" everyone would be sympathetic to them.

Instead they piss and shit all over about how their jobs are divine edicts from god and simulacras are demonic.

But part of it really is that making art is a very human thing to do, from the earliest records we have of humans, and mechanising it away with AI feels like chopping out part of the human experience. It isn't like "a better way to make cheesecake", where the AI is churning out industrial-recipe amounts in an industrial process. It's reducing creativity and imagination to a set of standard tropes for lowest common denominator appeal, like the production line of Marvel movies which, I think, people are beginning to get tired of because it's all too much and too the same: just slot in a new comic book character and sprinkle in explosions and fight scenes. A formula that gets over-used no longer works, because it's tedious. You've seen the same thing sixteen times before, why go see this particular one?

A lot of the complaining is taking themselves too seriously, but it's not merely about losing a job. It makes people feel replaceable, and in something that was considered to be uniquely human. Maybe a robot could replace you as a worker on an automobile assembly line, but as an artist? How would you feel to be totally replaced as a programmer, and whatever you might produce would be regarded as amateur hobbyist stuff, "that's nice dear", but everyone knows real coding is done by AI. Your occupation would be gone, and if this is something you do because you love this stuff, and not just as "well I gotta do something to make a living", wouldn't you feel lost and valueless?

It isn't like "a better way to make cheesecake", where the AI is churning out industrial-recipe amounts in an industrial process. It's reducing creativity and imagination to a set of standard tropes for lowest common denominator appeal, like the production line of Marvel movies which, I think, people are beginning to get tired of because it's all too much and too the same

(Emphasis added). I'm not sure where the bolded part came from. What reason is there to believe that AIs would reduce creativity or imagination to a set of standard tropes for lowest common denominator appeal? Nothing about the actual process of the creation of art by AI would imply that. If we look at usage of AI in other fields like, say, go or chess, AI has been known to display creativity far beyond what the best humans have been known to come up with.

I am old enough to have developed physical film in a real darkroom. Using negatives and developer was real work that took skill and helped build an appreciation for film photography. I think it was a really fun thing to learn, and I'm glad a did it, but oh boy is it useless now. Now, I can pick up my digital camera and have it automatically focus, adjust settings and snap off pictures at ~12 fps, apply all of the lens and color corrections and spit out gigabytes of jpgs onto a tiny memory card, then I take all of those photos and store them on my multiple-terabyte hard drive with backups in several locations.

I don't think the new-found ease in photography has rendered it inherently cheap, but has certainly opened the floodgates to a morass of shitty, low-level photography. I shudder every time I see a 'gram-girl (or boy) taking some basic selfie at a scenic location. But there is still a lot of photography to appreciate, pictures that take real work, not just in getting the right shot but in setting up the camera even now. No matter how smart the camera itself is, you still have to be at the location and looking the right way at the right time, and no matter how good it is at selecting a generically good setting itself, a skilled human can do better.

I installed Stable Diffusion a few days ago and, let me tell you, it's the real deal. My dumb, artistically-challenged caveman brain can put in 75 characters or less of generic prompts and in just a few minutes select from a slew of reasonably decent AI-generated art, select one and spend an hour refining it down to something I really like. It's the real reason I will be upgrading my several-generations old video card when the new 40XX series drops, not my habitual gaming, so that I can speed up that generation process. It’s absolutely blowing my mind, and I find it so very exciting to think about how I’ll be applying it to RPG or writing art.

But it’s still not magic. It has trouble taking very specific commands, it has trouble with anatomy, it has trouble with some prompts, it’s still limited in how many prompts it can handle, etc. It has a lot of limitations, many of which will most certainly go away with time, but for now I would liken it to having a decent artist who will immediately draw some art for you, but you can only communicate with them via tweets (less than that even!). Much of the skill in using it comes in through using other programs to clean up the images, removing artifacts and dropping the right “seeds” of implanted features for the program to take up. Another huge part is in getting a better feeling for how to give prompts and adjust settings to really get the most out of it. I expect the skill floor to raise up over time, so yeah, we’ll be inundated with reasonably good generated art.

This is already the case though! There are people who post on Imgur just dumps of elf art or pixiv manga art, endless seas of generic fantasy concept art, so much dross that fills DeviantArt with human-made but utterly indistinguishable work, work that people have pored so much time and effort into. Out of all that, only a few gems seem worthy of to keep around. How much worse can it get? I don’t see putting in prompts as really that terribly different now from entering terms in a search bar; that a human drew every line in one and not the other feels totally irrelevant.

I think the deeper question is a feeling about how much the AI is actually creating art, and how much of it is just "it got trained on a zillion images and it's just cutting and pasting according to your prompts; you tell it you want a buxom blonde woman in a bikini sitting on a beach and it selects out of all the stored images of buxom blonde women, beaches, and bikinis and trims it as you refine your prompt".

I think the fears and opposition come from a place where it's "a human imagines the work, puts it together, creates something new" while the machine (so far) isn't creating anything because it doesn't have a mind to think, it just does as it is told. Collage art, cutting out images created by humans and sticking them together in the combination you - not even it, itself, but the human prompter - tells it to do.

Do you get what I mean?

I think the deeper question is a feeling about how much the AI is actually creating art, and how much of it is just "it got trained on a zillion images and it's just cutting and pasting according to your prompts; you tell it you want a buxom blonde woman in a bikini sitting on a beach and it selects out of all the stored images of buxom blonde women, beaches, and bikinis and trims it as you refine your prompt".

If this is the deeper question, then it seems to come from just a fundamental misunderstanding of how AI art works. There's no cutting or pasting going on. Not unless you want to say that a human artist who develops his own personal style through observing pre-existing pieces of art and experimenting with what he can draw is just "cutting and pasting" from the images saved in his head from those observations.

My understanding of the mechanics of Stable Diffusion is very limited, but I don't think this "collage art" model is quite right. The computer doesn't really pull out whole chunks of images; it doesn't know what blonde, buxom or even woman are. But what it does have are statistical relations, so when it generates a bunch of noise it pulls out lines and shapes and colors based on those statistics, depending on the prompts, then makes a bit more noise on that drawing and draws again, and eventually it pulls a random-ish image from the noise. This reminds me of when I used to sketch, and I would lightly draw lines in pencil, then as the concept firms up you make your lines darker, until you're left with a fixed image that you can commit to pen. But that's all mechanical skill really; I can imagine scenes that I would never be able to sketch, much less bring to full art; the imagining and the art-drawing are separate to some degree. Is this better or worse? I don't know. But pencils and photoshop don't think either, and no one seems to mind.

But what has really made me ponderful is that the way Stable Diffusion creates art feels similar to the way I create art, and it appears to think somewhat how I think. Perhaps that's why its unrealistic mistakes go unnoticed sometimes, my mind fell into the same trap that it did and e.g. overlooked an extra finger, because my mind doesn't sit there and count fingers and neither does Stable Diffusion! It just takes what it sees and roughly maps the shape and position into the "hand" map and calls it a day. But even human artists have trouble with hands!

In earlier discussions on art AI, I expressed significant skepticism in AI generated art for two reasons:

  1. There was little proof that NSFW images could be done well, which was where a large part of online art commissions come from, and which indicated that there were likely some issues related to being able to get images of things you actually wanted instead of just taking whatever the AI would generate for you. Copying a human face in a portrait-style setup is one thing, but capturing bodies in various sexual positions without ending up with a cthuloid mess of dicks or at least falling hard into the uncanny valley is quite another.

  2. The lack of stability, i.e. that it was hard to create a character or theme, and then change little bits of it at a time, e.g. create an image of a person eating an apple, and then also being able to create an image of that same person sitting and reading a book.

The things I've seen on /hdg/ have pretty convincingly proven to me that issue #1 has been solved, or never existed in the first place. I've seen some pessimistic takes that it requires tons of time and 98% of it is garbage, but the fact that random anons on 4chan can generate the level of quality I've seen means AI art has advanced quite a bit more than I thought it had.

I can't believe autistic booru taggers became the heralds of an artistic revolution. Has anyone seen Gwern lately, or is he filling up a 500tb RAID with Asuka pics?

/hdg/: Is there an as109 embedding up yet?

Finally people asking the important questions

Has anyone seen Gwern lately, or is he filling up a 500tb RAID with Asuka pics?

Yes.

https://old.reddit.com/r/AnimeResearch/comments/xumxfk/novelai_diffusion_has_arrived_nai_launches/

Is the whole NovelAI thing worth dropping the ten bucks to try it out at the image generator tier?

If you like anime, then go for it. It's pretty good.

It's interesting, but if you're not hugely into this stuff, you can run a local StableDiffusion if you have 6GB+ of VRAM (sometimes down to 4GB, albeit slowly and with some frustrations) pretty easily using the automatic1111 UI, or it's possible to run on Google Colab for a couple hours a day at the free trial level.

NovelAI's version is better than almost all of its competitors -- I'm genuinely hungry to know how they increased the CLIP token limit -- but I don't know that it's 10 USD better unless you plan to use it pretty heavily.

First, I agree there is some value in putting in effort into something. It demonstrates a virtue in the person, the ability to delay gratification, to work towards a purpose. The reason we like this is probably evolutionarily determined, as such people are useful allies in bad times. Admiring people for the effort they put into climbing a mountain etc. is alright, it pushes us to become better and apply effort in smaller scale things. It's a symbolic distillation of our everyday struggles and shortcomings. That's all fine.

What I don't see as virtuous though is the other half of the attitude you show, namely that you want to feel that the stuff around you was done with a lot of human effort. Essentially this is the opposite of striving for efficiency, which we have been doing as humans since time immemorial. It would take more effort to swim to the other shore, but instead we build ships. It would take more effort to walk, but at some point people decided to ride horses instead and then invented cars.

Things that are made in an inefficient way for the purpose of demonstrating extra human effort are luxuries. Probably it would feel nice to be carried around town in a litter but why do that if there are cars? Understandably, it is a way to signal status if you can get many people to do inefficient work for you. Essentially it's a way for you to show that you can boss people around, having amassed (perhaps over generations) enough effort-tokens (presumably through some efficient method, using leverage, not by the sweat of your brow) to do this. It makes one feel important. I, however, think that the enjoyment of other people's senseless labor for showing off one's own status is a vice.

We should continue to use our brainpower to achieve more with less effort. This is not an argument to be lazy, but to work smart and get more done. Putting up artificial constraints makes no sense in general. Now if the constraints allow for the exploration of something interesting, that's another things. For example it could be a way to hone one's wits, eg the limitation of size in demo scene demos etc., to see novel ideas and creative solutions. That's all fine. It's also all fine if the actual hand made product is better. Furniture made of solid wood, designed to fit your rooms is better than the cheap stuff you buy at IKEA. But the reason to want it is that it's better. Also if you want lots of stuff done for you manually, how do you justify that? What makes you think that you deserve the fruits of all that effort? And independent of the answer, can you understand that many people can't afford having so many people jump around to their whim, and for them increases in efficiency can bring more improvement in quality of life?

Essentially this is the opposite of striving for efficiency

It most certainly is!

So I sat there and smoked my cigar until I fell into a reverie. Among others I recall these thoughts. You are getting on, I said to myself, and are becoming an old man without being anything, and without really taking on anything. Wherever you look about you on the other hand, in literature or in life, you see the names and figures of the celebrities, the prized and acclaimed making their appearances or being talked about, the many benefactors of the age who know how to do favours to mankind by making life more and more easy, some with railways, others with omnibuses and steamships, others with the telegraph, others through easily grasped surveys and brief reports on everything worth knowing, and finally the true benefactors of the age, who by virtue of thought make spiritual existence systematically easier and yet more and more important. And what are you doing? Here my soliloquy was interrupted, for my cigar was finished and a new one had to be lit. So I smoked again, and then suddenly this thought flashed through my mind: You must do something, but since with your limited abilities it will be impossible to make anything easier than it has become, you must, with the same humanitarian enthusiasm as the others, take it upon yourself to make something more difficult. This notion pleased me immensely, and at the same time it flattered me to think that I would be loved and esteemed for this effort by the whole community, as well as any. For when all join together in making everything easier in every way, there remains only one possible danger, namely, that the ease becomes so great that it becomes altogether too easy; then there will be only one lack remaining, if not yet felt, when people come to miss the difficulty. Out of love for humankind, and from despair over my embarrassing situation, having accomplished nothing, and being unable to make anything easier than it had already been made, and out of a genuine interest in those who make everything easy, I conceived it as my task everywhere to create difficulties. I was also especially struck by the curious reflection as to whether it was not really my indolence I had to thank for the fact that this task became mine. For far from having found it like an Aladdin, by a stroke of luck, I must rather suppose that by preventing me from intervening in good time to make things easy, my indolence has thrust on me the only thing that was left.

-- Søren Kierkegaard, "Concluding Unscientific Postscript"

"It's too difficult to meaningfully contribute to this society" vs "there is no difficulty any more, we must artificially make things difficult"?

Except for your example today weight lifting does have function. It’s the main means of maintaining a healthy human body.

Watching human weight lifting is a relatively niche sport especially for spectators but they still have ways to validate human versus machine lifting. Which brings up as said elsewhere weight lifters don’t argue for banning cranes. There will likely be a split between art made for production purposes - machine fine - and higher end art that has human provenance. We already do this with artifacts where being 10k years old gets higher prices than a modern replica that looks identical.

deleted

https://en.wikipedia.org/wiki/Lion-man

In some sense it's grading on a curve, but it's clear that substantial effort and technical expertise (for the tools available at the time) went into these prehistoric sculptures.

Renaissance works have been filtered and preserved based on quality. Prehistoric art probably went through a much more random series of events and what is left is probably not the pinnacle of the contemporaneous state of art but some random person's makings. But I have no qualifications to say either way, but it seems logical.

There was more than just a little improvement in the Renaissance state of the art, though. Perspective drawing, for the most extreme example, is probably the innovation that gives a painting enough verisimilitude in my eyes for me to really focus on what the painter was trying to depict without being distracted by the obvious flaw of distortions in the depiction.

As an aside, perspective also is a nice counterexample to the "image AIs were just trained on our work, that makes it plagiarism!" theory I see floating around. AI might be producing copyright-infringing works, but learning from other artists' works isn't proof of that. That's just how art works, which is why principles like one-point perspective went undiscovered for millennia only to then see universal uptake within a generation. Renaissance artists didn't all just suddenly get smarter at once (consider the delay before two-point perspective was discovered...), they were all learning from each other's works.

I can't imagine caring about the work that goes into art. Art is a product; the process is not something to exalt, it's something to lament, because the process has only ever been an imperfect means of translating the perfection of thought into the rough matter of reality with crude tools. A statue is beautiful because it is beautiful; it is no less beautiful if God willed it into being with the merest flick of his finger, and no more beautiful if the most devoted artisan spent his entire life on it.

Work for work's own sake is an abhorrent waste of time and energy. I rejoice at the prospect of AI obsoleting the artist -- and any lover of the arts should. A doctor should think a world that has no need of him is amazing; a soldier should long for a world that doesn't need soldiers; and wouldn't it be grand if we had no need of farmers, for all were fed?

The ultimate triumph of a field is self-destruction.

Art is a product; the process is not something to exalt,

The skill of it? The technique? The way the artist struggles with the medium, be it stone or paint, and solves for themselves the problem put to them? You see no difference in watercolours versus oils, impasto versus a smooth glossy finish, the way the artist blends colours and shades to achieve an effect - who would have thought that green and yellow and mauve would be part of painting flesh tones in a human face, but they are.

The invention of the camera didn't do away with art and the camera went from being "now we can have a perfect representation of a moment in time, no need for a painter to spend hours painting a portrait of someone when we can do it in minutes with better fidelity" to being used to create art itself.

It may well turn out that a lot of the current hysteria about AI art is just that - hysteria. That AI will become another tool for artists to use. It probably will replace a lot of commercial art, but maybe not - CGI did not replace humans as creators of images, now we need humans trained in how to use CGI to create effects.

The skill of it? The technique? The way the artist struggles with the medium, be it stone or paint, and solves for themselves the problem put to them? You see no difference in watercolours versus oils, impasto versus a smooth glossy finish, the way the artist blends colours and shades to achieve an effect - who would have thought that green and yellow and mauve would be part of painting flesh tones in a human face, but they are.

Those are all differences in the end product, too. The skill and technique are relevant so far as they create a different output to enjoy.

The ultimate triumph of a field is self-destruction.

Sublimation, surely.

Tomato, tomato. The ultimate triumph of medicine would eliminate doctors. The ultimate triumph of creation would eliminate creators. If we could all be our own little Gods, speaking worlds into being, it'd take a real odd sort to gnash his teeth over all the construction workers and carpenters put out of business.

Tomato, tomato.

mm.

If we could all be our own little Gods, speaking worlds into being, it'd take a real odd sort to gnash his teeth over all the construction workers and carpenters put out of business.

I'm inclined to agree, but it seems to me that this line of thinking butts up against some pretty serious philosophical questions about what we actually value. I'm not an atomic individualist, so a future where humans cocoon themselves away into perfect solipsistic selfishness doesn't actually sound all that hot. Making things easier is a good thing if the things in question are themselves good, but it seems to me that connection to others is something I'd miss if it were gone.

You don't need to be an individualist to see the value in trivializing the creation of luxury goods. Bond with people over a shared appreciation of the art, rather than the making of it -- which is typically a private and boring affair, anyway.

no disagreement there.

All of those are basically engineering to serve a purpose. Picture engineering exists, which is why AI art is even used. Picture engineering is how you obtain company logos and silly images for ads and presentations.

Art is a status competition. It sounds like what you're saying is art is dumb.

Art isn't dumb at all. I have plenty of pictures on my walls that I find aesthetically pleasing. They're not high status, they were free, gifts from a friend who liked Bos Ross landscapes, would smoke, and paint his heart out.

Art as an elite status competition is stupid, but I don't really care. Let the elites posture.

I agree. A parallel can be drawn between the invention of AI generated art and that of of optimizing compilers. They effectively replaced hand generated assembly code and I don't want to go back to a world where every programmer has to write their own assembly. It's much more productive to use higher level languages and reason about data structures and control flow, not registers and jumps. And just like AI art generators can make better art than the average person, a compiler can generate better code than even a competent programmer used to high level languages.

This doesn't make an understanding of assembly and the intricacies of the various families of processors any less useful in the narrower domain of compiler design. So maybe a keen understanding of art will be no less useful in a world filled with AI generated art, where the true connoisseur will appreciate the hand crafted art, while the masses will happily stare at machine created images.

Another comparison is procedural generated. I'm perpetually disappointed by it in games like Dwarf Fortress or the randomly generated side quests that games like Borderlands use - not because a human didn't put effort into each individual character name and backstory or quest goal, but because they're bland and boring for the most part, and it soon becomes obvious that even though there is infinite variety on the surface level, it all operates under a rigid structure.

a soldier should long for a world that doesn't need soldiers

Rather ironic given your flair!

This doesn't make an understanding of assembly and the intricacies of the various families of processors any less useful in the narrower domain of compiler design. So maybe a keen understanding of art will be no less useful in a world filled with AI generated art, where the true connoisseur will appreciate the hand crafted art, while the masses will happily stare at machine created images.

I like the technology, but I worry that this metaphor may not continue. In a world where computers just Did Everything for you in the programming and compsci spheres except the extraordinarily difficult exceptions, would many people be able to develop the interest or fundamental skills that eventually lead to a successful compiler design understanding? Or would their first genuine CompSci problem going to throw dependency hell at students that don't know what a file is, and they just turn around and say fuck it?

Sometimes it's chicken-sexing, sometimes it's a problem of available resource scaling (how do you train in your basement to run multi-million-user scale cloud?), sometimes it's a matter of developing the temperament to not throw computer monitors out windows. The First Step is a Doozy, and a lot of these skills are hard to learn and harder to understand what you have to learn.

It's plausible this won't happen. Past changes to CompSci haven't eliminated on-boarding opportunities, even if they've mangled many of them; artists have adapted structures to discourage bulk-scale tracing and seldom (have the specific tech skills to) take available 'traditional' automation tools to their maxima. Music is different than it was a hundred years ago, but it's not lacking steps for the garage band. Even if it does happen, there will always be the auties and the paranoid and the slightly nuts who hypnotize themselves into doing in the old-fashioned way.

And yet, there are skillsets that are lost, at least to country-scales. If you wanted to rebuild Saturn rockets pre-SpaceX, you'd have to start by rebuilding the entire aerospace industry (and might still today). There's a lot of woodworking techniques that have turned into gimmicks, shown only by weirdos on YouTube because you'd have to scour auction sites to even find the tools in the right quality to have a chance to learn the trick, because a trim router can do the easier variants and no one finds the hard ones worthwhile. There are classes of power transformer that can’t be made in IGBT or MOSFET forms due to physical constraints, and when the last guy who knows how to make the vacuum tube version retires, I doubt we build up a whole infrastructure to support a replacement.

And there's only so much confidence that induction can give you, in a case where the Type II errors are invisible.

I agree with what you say here. The way I see it, what AI is is the decoupling of art-as-status from art-as-product. People value the labor of prestigious artists because that artist has a monopoly on good artistic output. They respect the work so far as dedication is necessary to hone a craft, but if they could snap their fingers and make fantastic <music/video games/character portraits/illustrations for their novel/whatever>, they would, gladly.

Once the masses can produce art that satisfies aesthetic preferences, the status-artists will lose a huge market share. It's understandable why they'd bitch and moan, but it's the gurgling of a dying creature.

Rather ironic given your flair!

One must crush their enemies to enjoy the peace of a world without enemies.

I simply thoroughly disagree with this sentiment, and I am quite certain that I am not exceptional here. The process is important, the social context is important.

I would probably go as far as to posit applicability of some sort of labor theory of value: if you print out a random photograph, nobody will value it very highly, but if you paint the contents of the photograph on canvas, it will immediately be seen as having more value. Even more so, if we build technology that allows us to make a painting with a some kind of a gantry CNC painting machine, it’s product will be seen as less valuable than something that human painted by hand.

I think the above sentiment is shared by most normies, whereas your comment exhibits rather postmodernist ideals that few people actually share, as shown by revealed preference. Why are people spending millions on original artworks, instead of hanging cheap replicas that are exactly as beautiful? Because they strongly disagree with you.

I think most normies don't entirely care about the process. I could be wrong, but it really wasn't until recent decades where most people actually got to peer into behind-the-scenes stuff for things like movies, music, and video games. Now, there are definitely consumers and audiences of those things who do care and want to know, but at the same time, probably the broad majority of people in the world don't stop to think about how things are made, but just the thing in front of them.

Yeah, check out how few people view art streams vs viewing the same artist's art. Even very popular artists usually have less viewers than some no-name twitch game streamer or 2view v-tuber.

People into the process are mostly other artists trying to crib notes.

So what, specifically, is the source of that value? My intuition is that if the same painting were made not by a machine or a career-artist, but by a young child, or better yet, an animal, it would be seen as even more valuable, still. (Well, disregarding the effects of name recognition that can balloon chosen artists' work to staggering prices.)

At the risk of making things too meta, it seems to me like this value stands in proportion to how unusual (and thus rare and potentially otherwise-useful) the displayed skill seems to be. A great painter may be of extraordinary use in producing other great paintings that you want. A child prodigy, or an intelligent animal, may portend greater things still.

But a machine is just a machine, whose capabilities we know, just like we wouldn't care about the works of an animal if it were just an accountable product of instinct. The more of a good surprise it is, the more we treasure it, it seems.

But this is just my impression and I would be very glad to hear others with better theories.

I think you are in fact an exception, not the rule. The process isn't what matters to most, unless you happen to have an interest in learning about it. What matters is not the process, but the output.

And this isn't just for art. This is generally true. I personally enjoy programming and learning more about how different programmers have solved particular problems. I'm in the extreme minority, though. Most people don't give a shit, they want their software to do a task for them and don't care how it was made. A statistician may care about the beautiful mathematical model they use, but most people just want to be told the results. And so on, for pretty much any discipline you can imagine. People just do not care about the process by which things are made, unless they happen to have a particular interest in that topic.

Not to mention that the AI model itself is built using beautiful and creative mathematical ideas and engineering principles. The code can be elegant etc.

Why are people spending millions on original artworks, instead of hanging cheap replicas that are exactly as beautiful?

To show off. Jesus didn't say "yo, here's a loaf of bread, it's only one, so it's very valuable, make sure to hoard it". He instead multiplied the loaves of bread and the fish to feed the crowd. Sharing is good. If you have a reliable way to copy something, you should do it. Same way I think about file sharing, free software etc.

If what you say is true, there is no cause for concern! If there is some ineffable quality of realness to authentic human-made art, and revealed preference does indeed show people prefer it, then AI art is not a threat to real artists.

I wonder why all the artists don't have that same confidence.

I'd like to echo your sentiment and add another example to the mix, although not exactly analogous: I've worked for a few years on a programming project trying to compete with a closed-source service that has a monopoly. If, for whatever reason (it probably won't happen) they decide to open-source their work, I would be livid. To me, the wasted time wouldn't just be about a sunk cost and "no reason to finish," it's because the current WIP is itself pretty impressive, but it wouldn't seem impressive because for all everyone knows I just ripped off the open source version.

I can't really think of anything perfectly analogous to AI art, with how it retroactively invalidates past human effort:

  • engineering advances cheapen ancient wonders, but you can still appreciate how epic the Pyramids are because they didn't have a crane

  • things that used to be art (custom tribal weapons) can be manufactured when they have an actual purpose outside of just being art, but you can't mistake an heirloom for a mass-produced widget (Unless you're Anakin Skywalker building a protocol droid)

  • photography helped artists pivot to non-objective art, but it's unclear how a visual artist is supposed to distinguish their future work from AI-generated pictures, apart from only showing them in physical media.

EDIT: Would like to explicitly distinguish between some things here:

Motte: It's bad that you can't distinguish between AI art and human art.

Bailey: It's bad that AI art reduces the total status of artists.

The invention of cranes probably did reduce the status of powerlifters. In fact, I've heard theories about how technology is making brute strength obsolete in favor of knowledge work and symbol manipulators. That is, when the tribesmen saw big rocks on top of one another they thought, "how wondrous are our power lifters" and after the invention of cranes, they shrug and go, "yeah we can put rocks on top of each other."

If the future of seeing the average logo or image is "yeah our computers are pretty cool i guess" instead of marveling at how beautiful the art is, so what? As long as we can still run powerlifting competitions (where they aren't cranes) and also have curated museums and DeviantArts where somehow we could verify it was human created, isn't that all we need? Granted, that tagging technology might not be feasible.

but it wouldn't seem impressive because for all everyone knows I just ripped off the open source version.

...Is that not what credits are for? I mean, I'm not well-versed in crediting for open-source software that was made by more than one person, but still, you could probably at least put it on your resume.

engineering advances cheapen ancient wonders, but you can still appreciate how epic the Pyramids are because they didn't have a crane

This is exactly the situation with AI art, though. Nothing in the past is invalidated. I can appreciate the skill Van Gogh had to have, because he didn't have a computer program to help him. And conversely, I can still appreciate the skill that an artist today displays even if they could have used a computer program (but chose not to). There's no reason to say that AI art cheapens past, or even future, artistic endeavors in a way that engineering advances didn't for engineering endeavors.

The value that participants and spectators derive from the activity is purely a function of the amount of human effort and exertion that goes into the activity. Having a machine lift the weights instead would be quite beside the point, and it would impress no one.

But would you take the complaints of powerlifters demanding cranes not be used seriously? This is the crux of things, you want difficult art for love of the process itself, and if you want to join hobbyist communities to promote human created art for that sake more power to you. But to those of us who are not interested in this thing your complaints really do seem to boil down to making our lives worse so that people you feel kinship with can continue to monopolize a market. And it really does seem to be the market thing that is in contest.

Every piece of human art might be like Mt. Everest to you, and more power to you if you want to celebrate people who make that climb for no reason but leave the rest of us out of it.

edit: a question to cut out some speculation. If All ai art was reliably tagged in the metadata and software existed so that you could avoid accidentally seeing ai art and only frequent places where all tagged art was prohibited would that satisfy your complaint?

Cranes, like the cotton gin, manufacturing plants or programming language compilers, are engineering tools used to serve a purpose. That is, an actual purpose. Whereas things we consider art tend to be done because it is fun or for status.

The difference is, that the existence of a crane doesn't affect the status of powerlifters. You can still appreciate a power lifter because you know he's not a crane. To the extent that Stable Diffusion etc. mimic art, you can't really tell.

Now, there are a lot of good reasons to have AI-art generators. Like cranes, they can help us engineer and build things faster. People here have mentioned that AI art is probably already being used for generic business presentations for when a slide needs to be livened up and it doesn't need to be too precise or fancy for the audience to get the point.

Fine, artists no longer get their money ripping off people making powerpoints, but AI art still threatens the status market they're engaged in, which as far as I know, has no analogue.

Most artists (graphic designers) who get paid to do stuff on an everyday basis aren't the next Michelangelo and aren't doing something extremely novel at the forefront of artistic expression. They just design another corporate logo, paint weird Rule 34 images, etc.

Fine, artists no longer get their money ripping off people making powerpoints, but AI art still threatens the status market they're engaged in, which as far as I know, has no analogue

I think power lifting is a fine analogue. You can't tell if a building used very strong men to get concrete pillars in place or a crane and yet powerlifting as a competition persists. AI art is threatening the market potential for artists, it's not removing the ability to produce art the hard way.

I own a chess board and play friends occasionally, the existence of an ai that can produce superior play than either of us does not spoil our fun.

The invention of the crane reduced the reach of the powerlifter status market, because when people look at buildings, they're assumed to all be made by cranes, but you can still watch real people lift weights and they're obviously not a crane.

AI Art will reduce the reach of artists and their monopoly on making pictures. Maybe in the future, people will assume most logos and the like are made by computers. That's all well and fine. But how can you prevent imposters from submitting AI art to museums and competitions? It would be as if a bodybuilder could hide a hydraulic arm under his clothes (or take steroids!) and compete without working.

But how can you prevent imposters from submitting AI art to museums and competitions?

I'd lean the direction in how photography is also considered an art form. It's not the mechanical creation of the image that is relevant so much as the artistic intentionally and editing that results in the image to be displayed. Prompt engineering and selecting which of many different output artifacts seems rather analogous to composing the photo and selecting which of many (and for some many many) takes to exhibit.

I agree that promptmancy is the appropriate analogy to photography.

But I can't imagine prompt engineering is subjective though. Translating human intent into a prompt to make an image feels like a skill in a way that photography feels like art

But this just might be sloppy thinking on my part, or an opinion I've been socialized to hold

You keep bringing up the counter argument yourself. Steroids! I do expect high end art competitions will have to take a trick from the sports community and take some measures to prevent ai art from winning, they even have the advantage that one can't use ai at a live event like one can use steroids. I do think most commercial art will be assumed to have come from ai in the future, this will change the artistic landscape in real ways, but so what? Art status competitions haven't exactly been a great example for long lived consistency.

But how can you prevent imposters from submitting AI art to museums and competitions? It would be as if a bodybuilder could hide a hydraulic arm under his clothes (or take steroids!) and compete without working.

Same way powerlifting tournaments prevent imposters: make them do the labor on-site.

I wonder if that would really suffice for them. After all, the guy may generate the image at home on his PC, then memorize it and paint it on site from memory. If this is still "A-ok", then this is a weird esthetic preference.

That just sounds like how a classical artist might have memorized the way nature looked in a particular spot, then went home and painted it. If someone has the skills to do all that, I think they deserve the credit for it.

"This man didn't create art -- he held an image in his mind and put it on a canvas through the movements of his hand and arm!"

Bro, at that point he's not faking anything, he's just actually making art.

Is the art in the arm movements or the idea, the composition, the choice of colors etc? If I memorize how to paint a Mona Lisa replica, am I as impressive as Leonardo?

More comments