It’s not a moral judgement, that’s just how humans are wired. The lows make the highs higher.
I can imagine AI art having a similar effect (creating a glut of images/logos/whatever that devalues ones made with care) but am hopeful that we'll get better at filtering the cream of the crop. In 5 years tons of things will have AI logos that would have been made by a graphic designer (or simply not made) in the past. That sucks for graphic designers who are out of a job, is good for people who get cheaper logos, and TBD for overall society who now has lots more "custom" logos etc to wade through.
Obviously having roads is a great boon to a park's accessibility, and the ability of people with different mobility needs to appreciate nature. But it also made me thoughtful to imagine the feeling of wonder at seeing bridalveil fall after hiking for days into a roadless yosemite valley; how much more special and impressive it would seem after that journey?
This metaphorical tangent is pretty far removed from the original discussion, but how do you weigh the accessibility of a thing against how that accessibility changes its nature?
This is such an old man “I used to walk uphill both ways” take.
Not everybody has the TIME COST to pursue being an expert in art or code or whatever. But if they have an amazing idea and can now use AI to produce the idea then that is a beautiful thing!
For example: Having an idea for a cartoon used to be a dead end. It would die in your head because most people cannot stop their life and dedicate a substantial amount of time, effort, and sacrifice to produce the single cartoon idea.
What's the point in having an idea for a cartoon in your head if an LLM can just write an infinite amount of cartoon ideas in a heartbeat, and probably a better one than you came up with.
But drawning a cartoon isn't very challenging. Most of my peers could draw someone from South Park in a junior school.
The hardship in making cartoons is the amount of choices you need to make and the amount of knowledge how those choices would impact a viewer. If you delegate all of that, the cartoon wouldn't be simply blunt, it would be self-contradicting. And we already had a way of making cartoons, that allow your writing to shine through bland animation – since flash, actually. It might actually be even faster then using generative AI
This kind of mentality would ban Star Trek replicators, should they be invented one day. "In my day, you had to actually make things, we didn't get to replicate them, so we shouldn't, even if it's possible!"
Today's musicians have far greater access to lessons, recording equipment, inspirational material than 100 years ago.
Mountain biking (80s single speed with no gears, suspension, etc.) versus modern e-bikes with radial tires and hydraulic brakes.
Who cares? Value your own experience as you do. The less we all think about prestige, the more it will go away.
>The time is coming fast when the only living thing around a motion picture house will be the person who sells you your ticket. Everything else will be mechanical. Canned drama, canned music, canned vaudeville. We think the public will tire of mechanical music and will want the real thing. We are not against scientific development of any kind, but it must not come at the expense of art. We are not opposing industrial progress. We are not even opposing mechanical music except where it is used as a profiteering instrument for artistic debasement.
https://www.smithsonianmag.com/history/musicians-wage-war-ag...
AI generated images are only an extension of what e.g. photography has experienced in the last decades. We’ve had film cameras, then digital cameras, then smartphones, each of these commoditized image creation by a then-unthinkable factor.
It’s an ongoing process, even if this leap seems especially big.
I suspect the wannabees exist in the narrow window when technology has expanded enough for non-professionals, but hasn't seen wide enough adoption that the man on the street will recognize the pretentious self-aggrandizement.
No, I was talking about photography - and people replacing a digital camera with a smartphone. For most this substitution works very well; and the whole digital camera industry has shrunk significantly[1].
The photography community has been discussing wannabe photographers ever since my uncle bought a dslr and started taking photos at family weddings.
[1]: https://petapixel.com/2024/08/22/the-rise-and-crash-of-the-c...
You must not be looking very hard. There are many youtubers or influencers making indie films or shows.
NigaHiga, Annoying Orange and Shane Dawson all made movies. Freddie Wong started out as a Youtuber and created Video Game High School.
Go to Amazon and drop a few grand on mics, lights, cameras and lenses. The result is production quality beating any 90s talk show, which would have taken a whole team to do.
The trouble with things like climbing is there are only so many mountains to go around. We already can't walk in many places because of cars. I don't look forward to the day that similar vehicles can go up mountains. The existence of mass-produced clothing doesn't affect your ability to do your own knitting, though.
> The trouble with things like climbing is there are only so many mountains to go around.
This is taking the metaphor far too far. Nobody is literally taking mountains away from people.
There's a big gap between "pretty much nobody" and the reality, which is somewhere between one third and two thirds of everybody. You might want to reflect on exactly how your perception diverged so radically from reality.
Do people in cities suffer from not being able to grow their own food and make their own clothes? I don't know for sure, but official statistics claim that, even today, they commit suicide at much higher rates despite having much less material scarcity. Robinsonades have been a popular genre of fiction for centuries, suggesting that people long for that kind of autonomy. Today, we also have zombie apocalypse fiction, RPGs, and preppers.
From another angle, sports consist entirely of skills and challenges which aren't needed and never have been, suggesting that they don't get forgotten. Hobbies also consist of skills and challenges which aren't needed.
It's not about climbing.
is it a net negative to society if the average person could produce so much art that it becomes post-scarce?
It is enshittification.
We have no right to tell people they have to learn to climb to get to top of the Everest.
I can't draw but I want to create my Art using AI. What I now see is a bunch of people who associate their self worth with a rare talent and don't want others to join the party. I want to resolve the issues around copyright for training, but once this is out of the way I want to draw exclusively through AI because it's the only way I can do it. And I LIKE IT.
I'm a skilled pianist. The funny thing is that I heard similar criticisms about computer music a couple decades ago. "No playing skill needed". Despite knowing how to play, I'd rather do computer music nowadays anyway. Please stop telling me what I can and can't do!
Computer music, as it existed a couple decades ago, still played exactly what you asked it to, and it wasn't filling areas where you underspecified the music with a statistical model of trillions of existing songs. And that's the difference, for me: the ability to underspecify, and have the details be filled in and added in a way that to the audience will be perceived as intentful, but which is not.
Whether drawing (writing etc.) through AI counts as drawing (as making art) is a debate we have to resolve in the upcoming future.
As soon as we get more control over AI output, those arguments will finally die their well deserved death
To those that AI art offend: don't think of people as artists, but simply as art directors dealing with stubborn artists that won't ever work to spec
Dismissing the argument that we are losing something in this "democratization of creativity" by fighting a strawman that says you are not allowed to participate instead is a bit lazy
My, my, you really took the worst example to defend your point. The Everest is now an overcrowded dumping ground full of cadavers, shit and trash, with idiots putting not only themselves but their sherpas and other mountaineers in danger due to their arrogance, lack of ability and shittiness.
>What I now see is a bunch of people who associate their self worth with a rare talent and don't want others to join the party.
What I see is a bunch of people creating digital doubles of existing artists without their consent and using it to make money.
>Please stop telling me what I can and can't do!
Oh the irony...
You're just not shallow in the parts of reality that you care about.
Don't feel superior for you are not.
Unlike many, I am not going to follow along into caricatural post-modernism where nothing is good, nothing is bad anymore.
Explain to me why nihilism is factually incorrect?
Good and bad is all relative to the perceiver.
Code is a way of creating something that itself may or may not be good. But the actual code - I agree, it can be objectively bad.
The Nihilism concept, as I mean to use it here, is more about meaning, values, aesthetics. Concepts like Logic & Math, not so much.
Tangentially, what does enshittification mean now? Quoting Wiktionary, at one point it meant "The phenomenon of online platforms gradually degrading the quality of their services, often by promoting advertisements and sponsored content, in order to increase profits" (coined by Doctorow), but now people seem to use it to mean... things becoming shit?
You could argue this is the very sort of activity they were criticising when they posted! We are all vulnerable.
> Second: the fact that a neologism is sometimes decoupled from its theoretical underpinnings and is used colloquially is a feature, not a bug. Many people apply the term "enshittification" very loosely indeed, to mean "something that is bad," without bothering to learn – or apply – the theoretical framework. This is good. This is what it means for a term to enter the lexicon: it takes on a life of its own. If 10,000,000 people use "enshittification" loosely and inspire 10% of their number to look up the longer, more theoretical work I've done on it, that is one million normies who have been sucked into a discourse that used to live exclusively in the world of the most wonkish and obscure practitioners. The only way to maintain a precise, theoretically grounded use of a term is to confine its usage to a small group of largely irrelevant insiders. Policing the use of "enshittification" is worse than a self-limiting move – it would be a self-inflicted wound.
And, in some ways he was right... but also not. What eventually happened is that art education that would be required for any aristocrat or an aspiring aristocrat became optional. Most intellectual elite today would be unable to draw much beyond a stick figure diagram. Art, at least where it concerns drawing or painting became a narrowly specialized field. And so, most people today don't understand and don't appreciate art.
But, art didn't die. Instead, artists started asking themselves more questions about the nature and philosophy of art, figuring out which aspects are essential, experimenting more. In retrospect, it's sad that fewer people today have decent access or understanding of art, but while more common before, the understanding was often very superficial anyways.
Take a look at Super Auto Pets, a pretty successful and fun auto-battler game. It literally uses a free emoji pack for its core art. It doesn't really matter that they didn't hire an artist for those (though I think they did hire an artist after finding success) since a free emoji pack was enough for the creative product they wanted to create. If they had AI generated emoji instead, it wouldn't have really mattered much for the final result (creatively at least, I assume audiences would respond poorly due to GenAI's reputation). At the same time, the ability to create their product without paying a lot of money for artists was critical to make it in the first place.
This is what it means to me to "democratize creativity", to allow creatives to realize their creative ambitions in an area they are proficient at (e.g. video games) without requiring a lot of creative skill in adjacent areas that aren't critical to the experience they are trying to make.
If the art isn't critical to the game, then use simple art. It doesn't matter if the simple art is or isn't AI generated, what matters (in my opinion) is that it doesn't lead us into looking for meaning that isn't there.
However, complex art is needed to fit with genre tropes to attract the expected audience. It's like Apple shoving AI into their products needlessly—not a core part of the experience, but needed so Wall Street doesn't throw a hissy.
> looking for meaning when none is there
Or you generate your own meaning. Art is analysis for the audience.
No, people want to see their ideas come to life, previously this required effort in mastering a skill, now... it takes less and/or different amount of work.
That is the whole point of copyright, yes.
> no, you cannot copyright the Studio Ghibli art style itself. Copyright law doesn’t protect styles, techniques, or general aesthetics—like the hand-drawn, watercolor-inspired look with soft colors, detailed natural backgrounds, and whimsical vibes that Studio Ghibli is known for. It only protects specific, original works, like an individual film frame or character design from Spirited Away or My Neighbor Totoro.
And your answer... Doesn't seem to cover my question, I think. So the point of copyright is that the value of the style doesn't degrade?
It's just like an API. The very definition can be changed upstream by patches. People seem to miss that.
But here’s the real meat of the issue I guess, how far do we wanna go? At what point SHOULD something be un-copyrightable? What describes a certain “style”? My style is black&white, can I copyright that?
What would your diff be? :)
If we valued and supported artists without just seeing them as laborers, we could have a remix culture where no one really owns anything and no rights are reserved.
It's like pupils using an LLM to do their homework. They get the grade, but Idiocracy is awaiting.
Comparing systems like DALL-E 3 and Midjourney to modern illustration software is a fool's errand. Even with advanced graphic design tools like Photoshop you still have to employ the same artistic skills that have been handed down for generations: 3-point perspective, shading, basic poses, and sketching, etc.
If you can't draw, the fanciest graphic manipulation software isn't going be of much help to you.
As opposed to some doofus who has the ability to type in the words "3D", "trending on artstation", "hyper-realistic,", and "4K" and then proceed to churn out thousands of images in a single day. Stable diffusion is more like the equivalent of having your own personal artist on permanent retainer.
AI image generation is a gateway drug. You quickly grow resistance, and you will face withdrawal, too.
New tools and techniques that lower the barrier to entry always get over used, over hyped and bring in masses of people making everything from impressive new things to absolute cheap garbage. And then people learn the limitations, the tools get better and a lot of people stop using the tools because they realize there are still skills that go into the process and they aren’t interested in learning those skills. The tools find their place, and new masters rise into the new world. And then a new tool is created and we start the process all over again.
I bet a LOT of people would be an artist if there was money in it. Most don’t attempt because of the poor risk/reward of that profession.
I agree though that schools are even more in trouble now. But my point still stands: we have NO CHOICE. We must adapt, or Idiocracy awaits ;)
Schools definitely have to adapt, but there's so much inertia there, and in many countries, schools have a perverse stimulus to make as many pupils graduate as they can. It'll take another lost generation, I'm afraid. And the makers and exploiters of LLMs can take the blame for that.
In my view, the people who want to be dropped off at the apex of the mountain the most always dreamed of this outcome for a long time, even if only a little bit. They dreamed of the day they would be freed from the toil of having to study for years and years to produce art that satisfied their tastes, because they would not lower their tastes to make the process less stressful, or had other priorities so could not devote time to practice.
Now the market has innovated and their dream has come true. To people not serious about coming manual artists, there is nothing wrong with this picture. The market need is being fulfilled.
We have to ask where this market need to produce art at such a low level of effort comes from.
In your view, these people have always been cheapskates up in their minds, we just didn't realize it until their preferences were revealed by AI becoming available. If it hadn't, they just wouldn't be artists of any kind and we'd never understand they had any artistic ambitions at all.
At the end of the day, for whatever reason, these people want art they can call their own - it's just a matter of how much effort is required to realize their ambitions.
If these people are to reject AI, yet still care enough about creating art of some kind for a sustained period of time, you basically have to convince them that learning a hard artistic skill is more important than "doing the dishes"/whatever else occupies the rest of their time instead. Maybe a grueling 9-5 work schedule, for example. That is simply how the nature of practice/10000 hours-type advice works out.
If people for some reason just don't want to put in the time, but still want to produce quality art, then they'll choose AI. These two desires are no longer contradicting. They would have been 10 years ago, when you could just retort with "you're going to have to put in the effort, there's no other way." It's clear that that virtue of work-ethic being one's only path to results has been obliterated by AI, and to the new converts it sounds like gatekeeping in hindsight.
Cultivating new interest in learning a skill when it doesn't already exist is way harder than it seems. That gets into mental well-being and existentialist issues that many people in today's society find difficult to reflect about deeply.
I think capitalism has really done a number on people.
I mean, everyone knows it takes a programmer to spec an app. The "different" skill needed for vibe coding is regular old coding skillset. Only difference with image AI is that it doesn't do "vibe drawing" well.
AI art experiment is over. It's been long over, like camera based self driving was by the time some large orgs started embracing the technology. And it's taking longer for some to understand that it's over, just like that time.
No shade on stock photographers / illustrators, but that's where I see image generation end up at. And you could already get custom illustrations made for cheap on sites like fiverr. The real long term question will be whether an AI generated image can compete with services like that. (my guess: probably, AI images are higher resolution/quality and generated faster, but I don't know the real total cost of them nor that of cheap illustrators or stock images)
I don't think that rises to the level of "mass media has made the world worse", but it certainly has made it different and lost a few things.
That's exagerrated by movies. There were folk songs, but instruments - and players - have always been rare, and played by those who had enough time to learn to play - the poorest of the poor, the crippled and the blinds...
Note that at that time, either there was already a copyright to protect musical partitions, or if learning by ear the teachers had total discretion on who they taught and for how much. There were still barriers to knowledge.
And outside of Europe, any ceremonial/religious music was only taught to the initiated and played on rare occasions (there are interesting stories of Americans, etno-musicologists or even just musicians looking for new musical construction in the vein of Steve Reich, in the 60s, going to Africa to learn rythms with e.g. Nigerians, staying there for years, fully integrated - or so they thought - and bam! for a rare ceremony their hosts suddenly play something they never showed them before and refuse to discuss it afterwards)
Sorry, but most ideas aren't great. It's the painful process of refinement, hard work, and iteration that results in something great.
Many. And I love it when those people code something up with Scratch, Matlab, or even Excel. Even though I personally would dislike the aesthetics of the result, I’m much happier to see other people enjoying computers the same way I enjoy them (i.e., by using them to solve interesting problems) than if I insisted everyone do so using my preferred “real” programming languages.
Regardless of the tool used?
I don't think this is very true, I don't often see anyone bragging about their skills and demanding their outputs get put in a gallery and judged on equal merits as the old masters.
I'm not much of an artist and whenever I use an image generator to generate something, I don't do it to show off my artistic talent or whatever - if I was to do it before AI I would've commissioned an artist for it (which I probably wouldn't have done because it was too expensive, so I would forgo it) - the work the artist does would actually have even less of my own input than the AI's, since I'm giving them less description to go off - it's all the artist's, based on their own experiences.
Creative vision is not being realized, it’s being stunted.
There's nothing wrong with that.
People want to run marathons on shoes with padded soles and be called an athlete.
Anyway plenty of people can't even finish 1K, so 42.2K in padded shoes is still impressive.
If you wanna draw for the feeling of creative satisfaction that’s great. I just need an image on a page in service of something else.
AI is democratizing _execution_, not creativity. The advent of the smartphone camera era allowed for the average person to take vastly more photos than ever before, and yet the photos I personally find noteworthy are either remarkable in their context or aesthetic quality (e.g. [0]) or personally meaningful.
Alternatively put, "culture" values a proof-of-work that is creativity + execution. Taking a photo of a lunar eclipse setting over a crater has global value, because of the effort in planning and taking the shot (read [0], it's great). Me shitposting a meme for my game night groupchat neatly summarizing what happened in our last session of Pandemic: Legacy has local value, because of the creativity in adapting a meme format to our particular context and the execution of the edit, but nobody besides us would find it as funny.
(Interesting aside — I personally think that the photo in [0] is more impressive than someone going all that way to _paint_ the same scene en plein air, because of the split-second precision of the execution. Even though photography may have devalued a lot of the technical execution of capturing a scene, it has opened up new ways in which said execution can be valuable.)
So AI decouples creativity and execution in a way that is, like every innovation before, unlike every innovation before. The Ghibli aesthetic is the soft pastels of the color palette [1] and the comfortable, nostalgic character design [2] but also the restful rhythm of "ma" [3] and the detail lovingly lavished to background characters [4]. GPT-4o devalues some of those things, but it another sense it just re-weights our cultural proof-of-work valuations back towards human effort.
[0] https://lrtimelapse.com/news/total-lunar-eclipse-over-teide-...
[1] https://designmadeinjapan.com/magazine/graphic-design/the-ri...
[2] https://stylecircle.org/2020/09/the-studio-ghibli-closet/
[3] https://screencraft.org/blog/hayao-miyazaki-says-ma-is-an-es...
[4] https://www.rogerebert.com/reviews/great-movie-spirited-away...