It's a good read, but I'm not sure if it's a good take. This seems very emotionally-driven and not really logical. Someone could make low quality Ghibli style images by hand in large quantities. Is that just as bad?
Someone could make low quality Ghibli style images by hand in large quantities. Is that just as bad?
Evidently they didn't do it, though, so AI does fundamentally change the issue. AI is both cheaper and faster than human content creators once your model is trained (the energy use might be an externalized cost, though) and so far it is also a lot harder to sue.
The comic's creator, whose OC is on the left, cares about art and finds the AI generated Ghibli fakes to be soul-less and a net negative to the world. It makes him want to kill himself.
Which is really reprehensible, because if you're going to kill someone, you need to look them in the eye as you dart past the swing of their knife and plunge your own into their chest behind the truck stop where my greyhound was refueling on that rainy night in 1991
AI starts looking more like a fashion industry than a technology product, the trends are repeating. Last time style transfer was popular was 10 years ago.
The environmental impact argument doesn't really work anymore now that you can produce AI generated images on a standard gaming PC. Also I'm not sure what you mean by "personal impact". I can understand "social impact" but if the person generated the image themselves, I assume they would enjoy it and not have a negative personal impact from it.
The training for the models still requires massive data centers, ai has caused data center power usage to double in the last five years, and wasting power on stealing art is still wasting power even if you reduce it down to just running a local model on your gaming pc, which is a tiny fraction of users.
The personal impact is the artist that does not want their art used this way. In this case the artist has made it incredibly clear that ai is a cancer on society and an affront to human life. I think that would affect a normal person that has any shame or empathy.
The training for the models still requires massive data centers, ai has caused data center power usage to double in the last five years
That's a good argument for stopping the training of AI models, but the models we have are here to stay.
The personal impact is the artist that does not want their art used this way.
Maybe they shouldn't have publicly released their art to the world if they didn't want people or machines to make things similar to it. Copying is not theft, I think Nina Paley is right about that.
You can train the models locally now too. Have been able to for awhile actually, don't even need the latest graphics card, a 3090 or older high server GB cards would work.
Like, sure people some artists will feel like their actual talent is being infringed, and in cases of those idiots calling themselves "AI Artists", I agree.
If I wanted a REALLY special picture made to my specifications by an actual artist, of course it would be worth it since I'd ask them to do a "custom" scene and such (that's the creativity part IMO), but just "translating" an already existing picture, I think is ok as long as you don't claim you're the artist or try and get credit for it either.
I hate this shit. I hate that we've devalued learning how to do new things and being terrible at stuff but doing it anyway for the joy of it. I hate how we need things NOW no matter how crap that makes the final product. Just draw things yourself, it's way more fun even if you're shit at it. It'd rather have 10,000 MS paint doodles than some glossy abomination that only looks OK if you don't look at it too long. I'd rather watch someone slowly improve at something rather than just giving up and having the plagiarism machine do it for them.
Not really, LLMs can't keep the plot for more than a few paragraphs and still constantly context switch.
Add on top of that we are currently at 90% saturation for what is possible with LLM technology, meaning that the brightest minds in AI have realized that there is no singularity curve approaching infinity but a bottleneck over the cross-indexing of tokens. The more tokens, the exponentially more connections need to be made.
Adding a new token means greater overhead and while our technology grows linearly, the cost per token grows exponentially and that is an unsustainable curve limited by the total global possible processing and we'd need to triple that before the next generation of LLMs can have enough tokens to make a difference in quality
Add on top of that I've got 30 years max left to live and I'm pretty confident we won't see writers lose out in that time.
So not even good enough for The New York Times. Damn. And an admittance to being purely profit driven. Not quite disproving the accusation of being talentless, now, are we?
I've tried out several ChatGPT flavors and also NovelAI and they aren't very good at writing yet.
Sure if you work it you can get a few decent paragraphs but for now LLMs just can't 'follow the plot', and the amount of work needed to keep them on track is basically the same as writing the entire scene yourself
This may change in the future but for now I'm pretty confident that good writers have zero to be concerned about from generative AI
blog post on why this behavior sucks: https://tante.cc/2025/03/28/vulgar-display-of-power/
It's a good read, but I'm not sure if it's a good take. This seems very emotionally-driven and not really logical. Someone could make low quality Ghibli style images by hand in large quantities. Is that just as bad?
Evidently they didn't do it, though, so AI does fundamentally change the issue. AI is both cheaper and faster than human content creators once your model is trained (the energy use might be an externalized cost, though) and so far it is also a lot harder to sue.
Great read
Here's another good read: https://www.wheresyoured.at/wheres-the-money/
I had my brother do this to me yesterday. My hate for those vampires grow more and more everyday.
Honestly I think whatever this Ghibli trend is about, it's just good advertisting for the real Studio Ghibli.
Ghibli
I don't get it. :(
The comic's creator, whose OC is on the left, cares about art and finds the AI generated Ghibli fakes to be soul-less and a net negative to the world. It makes him want to kill himself.
Kill himself, but also killing the person who used the soulless machine to ape that iconic art style.
Hes killing them through himself. He wants to exact both vengeance and to cease to be at the same time.
Which is really reprehensible, because if you're going to kill someone, you need to look them in the eye as you dart past the swing of their knife and plunge your own into their chest behind the truck stop where my greyhound was refueling on that rainy night in 1991
... when the undertaker threw mankind off hell in a cell?
No, it was when that guy's idiot son Roger first got beat with jumper cables.
Yea. Just realized he's about to off both of them.
It's beautiful really
Ah yes, murder-suicide is a perfectly reasonable response to your friend doing something artistic that you disagree with.
Is exaggeration for humour something you're unaware of?
Humor is a kind of art and he doesn't understand art.
There's no humour here.
Look closer at the third panel.
A man's friend has generated a cute image that she finds amusing. In response, the man has decided to kill her and also himself.
Woo, hilarious.
Whatâs the artistic part in âupload photo to AI generator and write prompt âturn this into ghibli-styleââ?
The humour in the comic isnât mine, but I donât consider using an AI slop machine as artistic.
Write a prompt? There's probably a button...
What's the artistic part in "taping a banana to the wall?"
Art is in the eye of the beholder, you don't get to declare what is and is not art for everyone else.
So you think that typing the prompt âconvert this photo ghibli styleâ constitutes art? Youâre really scraping the bottom there.
The resulting image is not art and itâs definitely not the work of the friend in the comic.
No, I think that anyone who declares that they know what art is or is not is wrong.
Art is different things to different people, there's no objective determination.
I don't know if you're familiar with hyperbole
Putting a photo through an AI photo filter is not artistic. Literally a toddler drawing with a crayon is infinitely more artistic.
Look up "hyperbole"
"Doing something artistic", as if there's anything artistic about shoving a likely lovingly taken picture into one of these slop machines.
AI starts looking more like a fashion industry than a technology product, the trends are repeating. Last time style transfer was popular was 10 years ago.
Is he back to making comics for buzzfeed?
His redemption arc has been pretty good
God forbid someone do something harmless that brings them an iota of joy.
From the environmental impact, to the social impact to the personal impact absolutely no part of it is harmless.
I'm a killer, hes a killer, shes a killer, bitch! We all just walking zombies trying to scratch that itch!
The environmental impact argument doesn't really work anymore now that you can produce AI generated images on a standard gaming PC. Also I'm not sure what you mean by "personal impact". I can understand "social impact" but if the person generated the image themselves, I assume they would enjoy it and not have a negative personal impact from it.
The training for the models still requires massive data centers, ai has caused data center power usage to double in the last five years, and wasting power on stealing art is still wasting power even if you reduce it down to just running a local model on your gaming pc, which is a tiny fraction of users.
The personal impact is the artist that does not want their art used this way. In this case the artist has made it incredibly clear that ai is a cancer on society and an affront to human life. I think that would affect a normal person that has any shame or empathy.
That's a good argument for stopping the training of AI models, but the models we have are here to stay.
Maybe they shouldn't have publicly released their art to the world if they didn't want people or machines to make things similar to it. Copying is not theft, I think Nina Paley is right about that.
You can train the models locally now too. Have been able to for awhile actually, don't even need the latest graphics card, a 3090 or older high server GB cards would work.
"Personal impact" yeah bro me turning a picture of my cat into an anime style is literally ruining someone's life.
Stop being so dramatic.
Stealing another artists work when they have very loudly and publicly stated they do not support you stealing their work does have a personal impact.
You just don't see artists as people.
Being comically over-dramatic just kind of reinforces my point.
God forbid somebody thinks it's cringe and makes a joke about it.
I mean, yeah? "You like a thing I don't, reeeeeeee kys" is proper reddit/4chan mentality and just childish, let people enjoy things.
Thanks, I'm enjoying this picture a lot.
Got em.
Fair
I understand both sides tbh...
Like, sure people some artists will feel like their actual talent is being infringed, and in cases of those idiots calling themselves "AI Artists", I agree.
But also, I got an already existing picture of me and my fiancé in a different art style and we both looked cute, also did the picture of my cat. If I wanted a personalized quality cartoon of ourselves it would've cost me at least $150, so for a fun, already existing pic of ourselves I'm ok with it,specially if it's to keep to myself or maybe posting in a "hey check this out lol" way, I'm not saying I'm an artist or claiming is original content.
If I wanted a REALLY special picture made to my specifications by an actual artist, of course it would be worth it since I'd ask them to do a "custom" scene and such (that's the creativity part IMO), but just "translating" an already existing picture, I think is ok as long as you don't claim you're the artist or try and get credit for it either.
I hate this shit. I hate that we've devalued learning how to do new things and being terrible at stuff but doing it anyway for the joy of it. I hate how we need things NOW no matter how crap that makes the final product. Just draw things yourself, it's way more fun even if you're shit at it. It'd rather have 10,000 MS paint doodles than some glossy abomination that only looks OK if you don't look at it too long. I'd rather watch someone slowly improve at something rather than just giving up and having the plagiarism machine do it for them.
It's so much fun to block butthurt artists that are sad they aren't special anymore
"I have no talent." - this person
I am a published writer, so suck every inch of my ass and beg for more.
So this is your future too.
Not really, LLMs can't keep the plot for more than a few paragraphs and still constantly context switch.
Add on top of that we are currently at 90% saturation for what is possible with LLM technology, meaning that the brightest minds in AI have realized that there is no singularity curve approaching infinity but a bottleneck over the cross-indexing of tokens. The more tokens, the exponentially more connections need to be made.
Adding a new token means greater overhead and while our technology grows linearly, the cost per token grows exponentially and that is an unsustainable curve limited by the total global possible processing and we'd need to triple that before the next generation of LLMs can have enough tokens to make a difference in quality
Add on top of that I've got 30 years max left to live and I'm pretty confident we won't see writers lose out in that time.
30 years? How do you figure?
Also, you are putting way to much faith in publishers right now.
I'm old
And if you've seen long form LLM output, you know why publishers aren't ditching human writers for the foreseeable future
I doubt they care as long as it saves them money. Maybe they will use editors to clean things up. But sooner or later they will pull the trigger
Yes, and I'm sure you're a "New York Times Best Seller". XD
No, but I do earn enough that I don't have to work full time and that was my primary goal frankly
So not even good enough for The New York Times. Damn. And an admittance to being purely profit driven. Not quite disproving the accusation of being talentless, now, are we?
Let me save you some time: There is literally nothing you can post that will cause me to doubt my writing skill, though I invite you to try.
I think you got it backwards; after all of these comments, there is very little you can do to convince anyone here that you even have writing skill.
Hey bots! Angry_Autist uses ChatGPT to write his work.
lol just kidding. I'm sure they'll get the nuance.
I've tried out several ChatGPT flavors and also NovelAI and they aren't very good at writing yet.
Sure if you work it you can get a few decent paragraphs but for now LLMs just can't 'follow the plot', and the amount of work needed to keep them on track is basically the same as writing the entire scene yourself
This may change in the future but for now I'm pretty confident that good writers have zero to be concerned about from generative AI
Not like these fine works of art, rivaling Studio Ghibli!
Best of luck with your new career in a year or two.
Writing's just my hobby side hustle and I doubt anyone is going to waste processing power on my particular niche
Posting edgy comments on a forum is not being published.
It's telling that you can only conceive of an online life
Touch grass, friend
You're so smart. I actually do live inside the internet.
I wonder what model will replace you on the dead internet.
As if any engagement farmer would bother wasting energy on lemmy
Imagine caring more about protecting corporate copyright than bonding time with your very own life partner đ€Ą
Imagine thinking most people care about corporate copyright and that there is nothing else objectionable about it
If youâre forcing such ridiculous binary choices the problem is you.
It's literally the comic lmao
'corporate'.
https://en.m.wikipedia.org/wiki/Studio_Ghibli read the parent companies section