Expect a massive marketing reversal in 6-12 months, when every "AI-powered" company removes the keyword from their home page.
It has always been like that in arts. Back when I was a teenager in the late 90s, my art teacher was a 50+ year old guy that never grew out of his hippie phase and was absolutely non tech-savy. Were tasked to paint a sneaker onto DIN A3 paper, and find our sneaker model from mailorder catalogues of shoeware as a template. Now, the catalogue images were far too small to do a 1:1 copy. But tech-nerdy me had a 486 PC with a scanner and an inkjet printer at home - I scanned the image from the catalogue, scaled it to 2 DIN A4 papers and printed them individually. Then just put them underneath the DIN A3 paper and copied it 1:1 during art classes. Of course I ended up with the greates, most precise image. The teacher took the painting and praised it in front of the class, giving me an A. All this admiration and attribution from my classmates would have gone down the drain, if I just told anyone how I did it. That is basically the same concept - it was the predecessor of AI that I used, and revealing it would have made me a fraud.
s/made/revealed/
Revealing it makes you known as a fraud.
You realize that the admiration and attribution wasn't real, because you didn't actually do what you claimed? I get that you wanted to feel the effect, but that doesn't mean NOT revealing that you didn't do what you were being praised for made it real.
Edit: my point is different techniques are fine in different contexts
The revealing thing is how the cheating is seen as normal and how the other people are considered out of touch or wrong.
Tracing is a cardinal sin.
People lose careers over not understanding the massive difference.
Philip Steadman, Vermeer's Camera: Uncovering the Truth Behind the Masterpieces, Oxford University Press, 2001
lol! /s
Edit: I can also see the use of tracing with very small children or beginner adults with poor motorics to develop muscle control, to learn how to control the line weight and variation. It is really ineffecient when learning how to see and understand the 3D forms, though.
I didn't know if generative ai has a strong enough signature to go through the same evolution, but I wouldn't be shocked if it did.
Nah, they were still keen to let people know it made you a clown.
AI is literally the exact opposite of that. We had been building software with ruler-like precision even when it was unnecessary and got in the way. That's the only reason we ever wanted AI to begin with.
One mistake people keep making is the assumption that arbitrary and probabilistic imprecision ("wisdom of the crowd") is automatically equivalent to artistic or experienced/curated choice. It's absolutely not. Another mistake is that approximations are always enough to satisfy all problems, or at least "real world" problems. That's obviously not true either. That's flat out ignorance.
What I’m saying is he made an explicit artistic choice to use AI and it’s an integral part of the work. You can’t tell a creative person they aren’t allowed to do something.
AI is going to be used, and already is used, by creative people for legitimate uses that only enhance the statements they are making.
The fact that lazy people are using it as a shortcut for a lazy audience is not an indictment of AI
I'm saying that less intentional use has less artistic merit. AI hasn't made an impact on the toil involved in the creative process. It's instead another brush.
It all boils down to the fact that humans are not outsourcing anything to AI. If anything, it just raises the bar and makes humans have to work harder. Decision fatigue was already a problem long before this era of AI, and AI can make it worse.
That's exactly what happened to music with beat quantization and autotune. People just zoomed in, nerded out, got snobbier, stopped seeing the forest for the trees, and ultimately produced very boring music from exhaustion. Not only that, but a ton of nuance was lost. We took so much agency away from the performance and made it all march in line and in time. It's up to the audience where the breaking point is.
Exactly. AI generates imitations of the past. So does a paint brush at its fundamental.
That sums it up for me. This "new thing appears, become hot, gets overplayed, people hate it" cycle happens over and over, but that is trendy norms. Meanwhile, creative use will continue to go up and to the right as people learn how to better wield AI to express themselves.
The future won't be "AI in art is bad" but "AI as a tool of artists is powerful, AI as an artist is nonsensical".
AI is still in the 2010 autotune phase, where more is better.
Though if 2010, the year that Lost in the World was released, is your marker of an awful era, then I suspect that when generative ai does get used to create real art you will likely not be a fan of that either.
I could come up with a thousand examples, but I remember when I was in college and listening to a lot of 90s music. Blink 182 had this album come out, and I couldn't even listen to 3 seconds of vocals because of the insane amount of pitch correction grinding the life out of everything:
Or 22, A Million by Bon Iver.
What can generative AI do that can't be done, and done better, otherwise? Nothing.
It's not an attempt to replace or add an artistic medium; it's an attempt to replace the artist. It doesn't open new avenues of expression, and it has negative journalistic value too.
People who "make AI art" are not making art, nor making anything else. They are commissioning images/videos/text from a bot, just as people commission art from humans. The amount of praise/admiration this deserves is zero.
The unfortunate truth for art is that gen AI will stay, and they will take work from artists, just like photography did from painters. There will be a niche for it at some point, and until it finds that niche people will fight over it.
This feels similar to the timeline that lots of new technologies go through. Blockchain (put that into everything!). The internet + dot com era were probably similar as well.
Treating ANYTHING like an everything tool just seems to feel this way. There's no reason I want to use AI to buy something on Amazon. Or plan a vacation for me. I absolutely want to use AI to scan through 5000 files and find that one thing I was looking for though. Different applications.
You are using Claude like people use search engines. Back then there were tons of people who said they don’t need the internet or google to buy things just like they say the same things with AI
But people don’t control the market. Once Amazon bullied their way to 50% of all retail sales people no longer had a choice. It’s all coming from the same warehouses so you might as well get the lowest price
If it’s cost effective to have an LLM operated business, that business will outcompete the ones you like until they either fold or end up looking like everybody else.
With your personal agent, you can use it for everything YOU want to do and I can use mine for everything I want to do.
But we both log onto Amazon and are greeted by Rufus, a stranger that I’m sure most people have no interest in using.
And maybe shopping is a bad example, but AI is getting jammed into EVERYTHING and I’m sure that’s not the end result.
AI + a custom search engine - which might not use Google's metrics, or Google's sponsorship - would be a killer product.
So my prediction is that a lot of aggregator and middle man sites - like Amazon - could disappear within the next decade.
And Google might too.
An open network of independent search agents would be a gamechanger for the entire Internet.
Which is not to say this would be easy. But AI is actually well-suited to making ideas like these possible.
It would - finally - be genuine disintermediation.
Historically, including "AI" in a name meant it wasn't complete or ready for prime time use.
I'm not convinced our AI nor ML implementations that throw "AI-powered" into the marketing copy have meaningfully changed that notion.
We don't call spellcheck "SpellcheckAI" or "SpellcheckML" but it is arguably one of the best integrations of AI we've ever seen.
I think AI is there to stay for corporate media. The mental image of what is art and what is just a product for a company will just have to change.
Imagine how portrait painters must have felt when cameras rolled around. Yet there is a place for painting and for photography, both in art and (at least until now) in corporate media.
As long as people lie about using a tool and claim they did it with the high effort way, this won't be able to settle into a new reality without people fighting about it.
I think there will be a place for AI in art too - just not in doing the same thing as current artists are doing (and expecting the same praise for less effort), but there will be a niche for it too.
I didn't end up buying at that point, but 6 months ago my existing phone failed, so I had to return to the shops. Same phones for sale more or less, but not a single word about AI anywhere.
I wonder if any movie special effects that you notice will be considered AI?
Zach King is a YouTuber to watch on this topic.
The slogan that keeps coming to mind to me is 'AI has a massive PR problem'
When such a disruptive and powerful technology is made popular by slop art that undervalues creators and shits upon the cultural value of the arts, by unlikeable billionaires telling people how you'll soon be out of a job, by having it forced upon you by your boss, by marketing teams pushing for doomerism to increase the IPO value, you know that no PR campaign will ever save it and make it palatable for the masses.
And it's for the best, I'll say. AI as a technology is here to stay, but the masses will be wary and mildly disgusted by it for a very long time. The lunatic dreams of artificial superintelligences ruling over humanity are accepted to be blasphemy, not a glorious future to rush towards. The tech world is still too far deep in its echo chamber to see that no one really wants whatever they are selling.
That’s because it’s not your work.
It only takes a single project to show it works to get other people jumping on board. Some amount of gamers are pretty vocal about it (though I'm sure it's being used in most studios these days in some fashion). If a company creates a game which embraces the "slop" and has great assets on a fun game are people going to avoid it? If that works out, will companies following them avoid it? That's just not how people work.
But that is NOT the selling point of AI in relation to art. The selling point is "it will make all artists obsolete and unemployable". Literally. That is how Sora was sold too. It was not sold as tool for artists. It was sold as replacement for artist.
If AI ends up as a tool for artists, then that tech failed on its creators terms.
And quite a few on HN run that talking point too.
> Was it not marketed as a tool which enabled people to do more?
Oddly, no. I remeber it, because I found it weird. I found the open hostility toward creatives very noticeable. The way this campaign came across made me reject the "refuse to engage with modernity" style of arguments. If you run around insulting and taunting people, them rejecting you is not necessary rejection of modernity. It is rejection of insults and taunts targeted at them.
But also people have bad taste. AI music really only works on people who don’t critically listen to it. Most people just seem to want a catchy hook and the general structure of a pop song, and that’s how Spotify gets away with it.
But you ask people to pay even $1 for a track they will listen to hundreds of times? No thanks.
So people get what they deserve.
> But you ask people to pay even $1 for a track they will listen to hundreds of times? No thanks.
People used to buy albums or listen to the radio for free. If anything, paying Spotify means that quite a lot of people pay more then before for music ... however it is spotify who benefits rather then artists.
Just like with studios which used to have captured market and earn more then artists themselves.
Indeed, and we're now seeing similar hostility towards mathematicians, of all people. It's disgusting, honestly.
For example, creating a game where before you'd have to pay for assets or creating a movie to where you'd have to pay for a lot of various things. I gave an example earlier where I've watched vids people generated based on existing fiction. These are artists (you're still framing the shot, selecting dialogue, etc), and they could not have done this before; it would have been too expensive.
Some artists get cut out of the loop but it allows small teams to do what was impossible. Is this not valid?
And maybe film is different in some respects, but AI does not allow musicians, artists, or writers to do anything they could not do already. The only benefit is churning out mediocre content faster.
The animated movie Flow won an Oscar, and that was made by a very small team and entirely in Blender.
Fanvids and AMVs made by clever sampling and splicing have been a thing for a very long time, too.
There are also a lot of games that exclusively use free or self-made assets.
Money allows access to better tools and bigger teams. But replacing artist jobs with AI doesn't seem like a great way for artists to get money (or time to focus on art, for that matter).
Generative AI, especially for video, also keeps getting more expensive, as AI companys' bills finally come due.
Just iteratively refusing what a stochastic generative algorithm presents to you doesn't make you an artist. It makes you a client with attitude.
* https://en.wikipedia.org/wiki/Puppy_(sculpture)
* https://en.wikipedia.org/wiki/Michael_Jackson_and_Bubbles
etc.
Koons didn't write a two paragraph description of his ideas and then handed it over to a porcelain factory that repeatedly tried to come up with something that fit the spec until Koons cried out "Yes, that's it! That's what I had in mind!"
He drew sketches, built models, shaped prototypes, etc. He was an active participant in the creative process from the very start to the end, not just a client that checked in on what the artists were doing from time to time to give the thumbs up or thumbs down.
Pope Julius II did not paint the Sistine Chapel's ceilling. Michelangelo did. I cannot believe I have to say this.
If just "being involved" in the "creative process" is enough to be called an artist, then by that measure that includes the guys that are simply involved by having the money to patron the process. Maybe it also includes the guy delivering the pizza.
But I think you know that and you're just straw manning.
edit- to be clear i'm talking about using existing software like logic pro, blender, etc. i'm not talking about generative AI.
Just selecting from a random sampling makes you a curator at best. You know where you want to go, but don't know how to do that. If luck is not in your favor, you will never arrive. And you're wholly dependent on luck to get there.
Having never learned the computer tools, like Logic Pro, or whatever open source alternatives are available, I've been considering taking a crack back at it. I have no desire to learn these tools, but if I can say "ok move the kick 1/8 and double up the snare on measure 4" I may get back in the swing of things.
Forced retirement at 50 has me looking for new hobbies.
I'm not really worried about what anyone wants to call me, but I was just pointing out a legitimate use case for agentic AI in the arts.
God I hope so! Modernity sucks.
On the one end - somebody writing a book alone with a pen and paper, about an idea they had in a dream. Whilst they didn't make the book from nothing (somebody else created the pen and paper, somebody taught them how to write and structure a story, they are just using them to write a book) This is very clearly the work of that one person.
On the other extreme, imagine a wealthy person hires a writer to write a book. They specify that it should be about unicorns but nothing more. Let's imagine that they haven't even read the book yet. Can the wealthy person claim "I wrote this"? Most people would say no. They caused the book to be written, but didn't actually write it.
In my mind, asking an LLM to create something for you is more like the second example than the first, but there isn't a hard line between those two. I can think of plenty of examples which sit between the extremes (spellcheck, proof-readers, collaborative projects, re-tellings of existing stories). LLMs are in this space. You could have an idea for the book and lay out notes and chapter outlines and more, and have the LLM bring it together. Did you write the book? Who can say? Investigate it hard enough and the question becomes meaningless.