Art is truth, but tech CEOs art only resembles their truth - profit. Art is a gift freely given, thus degrades when profit is painted into the picture.
We are clearly riding a wave in which hypocrisy is virtue, lies are truth, gross-outs are charming, and believing in human dignity or creativity is reviled as a sucker's bet.
What's especially sad is how this mentality invites losers to double down on their baseness.
Only someone who already felt like sad-sack would find a race to the bottom appealing. One day these people will wake up and find that their jokes are unfunny and everyone else decided long ago that follow-the-leader is a bore.
Maybe it won't be too late for them to get a life.
Imagine promoting the fact that you stole Studio Ghibli's intellectual property. Altman is waaay too comfortable, but I hope it will be OpenAI's downfall.
In my experience as an artist I've found that most AI users do hold artists in contempt, I've had so many people be outright rude and patronizing when I point out how artists are being impacted by this tech. Their favourite word is "cope". They could just be a very loud minority (as is often the case with the Internet in general) but man are they LOUD. 😒
Mine too, and that's the assumption I made as well — always giving folks the benefit of the doubt. But they're at least both loud and powerful, if not particularly numerous; so a threat to be sure; and certainly dominant in the industry.
"Fair use" normally involves using *very small excerpts* of a work, for purpose of commentary on it. Not systematically digesting every pixel of the corpus.
I feel nothing but sadness about this. I guess that makes me human still.
AI slop will continue to devalue art and, yes, we will keep moving onto the next shiny thing that distracts us. What we’re doing is destroying culture one prompt at a time. There is no ‘but’ that fixes this, no matter what the snake oil companies tell us.
AI should not be shortcut to human artistic creativity.
The resentment and disrespect of artists and their craft by these reactionary posters does nothing but vindicate people's already existing stance against AI. For all of their talk of "democratizing creativity", all they ever seem to do is punch down at creators who are already struggling during these hard times.
When did Altman change his X avatar? Was it immediately upon rollout of the new ChatGPT image generation features, or in response to the user-driven Ghibli trend? That timing matters. From the tone of your piece, it sounds like you lean toward the interpretation that Altman led and users followed—but does the evidence support that?
In testing the new image generator myself, I’ve found the safety filter to be surprisingly aggressive. Even seemingly innocuous prompts can trigger censorship for violating "community guidelines" or "safety standards." Applying the Studio Ghibli style appears to function as a workaround—it allows content through that would otherwise be blocked. I explore this in more detail here:
"Draw a comic strip image in the style of Spirou and Fantasio where both characters sit in a red sports car, driving out of view into a scenery of open country. Chasing after them are several dogs. The car leaves a cloud of dust behind it. The sun starts setting and there are scattered clouds, some birds in the air."
Answer: "Here’s the comic-style illustration inspired by Spirou and Fantasio, with them driving a red sports car into the open countryside while being chased by dogs. Let me know if you’d like any adjustments!" Well, ChatGPT drew a Disney cartoon style image with Goofy-like characters in a car, where the car is chasing the dogs, not the other way around. I pointed that out to ChatGPT:
"Sorry, but this is American style 1950s not Belgian style 1960s comic strip. Also, it's the car chasing the dogs, not the other way around. Can you correct that?"
"Here’s the updated illustration in a Belgian comic strip style, now featuring Spirou and Fantasio in their red sports car chasing after the dogs. Let me know if you need further refinements!" This worked, the characters and the car looked for more Belgian-style. But out of five dogs, only two were chasing after the car, the other were in front or at the side.
Me: "Better. The dogs need to be chasing after the car, not going in front of it."
ChatGPT: "Here’s the revised illustration—now the dogs are correctly chasing after the car instead of running ahead of it. Let me know if you need any more tweaks!"
Me: "I'm sorry, but there are still two dogs going ahead, one after, and one beside, also two in the car. So of a total of six dogs, only one is actually chasing after the car. And the comic figure is again reverted to 1940 US comic strips. Did you forget my original instructions?"
ChatGPT: "It seems like I can’t generate any more images right now. Please try again later. Let me know if you'd like me to refine the description further so it's ready when the tool is available again!"
I'm really sorry, but ChatGPT should be called CheatGPT. Or dummy-dummy.
Parasitic behavior, the hypocrisy is disgusting, but... on the one hand I possibly couldn't eat as much as I want to throw up about the marketing. On the other hand, how is this new feature different from using text written by humans that did not give their consent expressly for LLM use and appropriation? Any LLM user, myself included, can't really throw the first stone.
Well, everybody pees into the swimming pool, but when you do it from the three-meter board more people tend to notice. We as societies seem to lack the appropriate decency or (legal) means to prevent freeloaders from succeeding. During this presidential term nothing will change. Let's wait for Musk's dressing down of "cry babies".
It's evil towards Ghibli, but what for? OpenAI or any their customer could not produce complete new movies "in the ghibli style" for profit, that would be quite an IP violation, and anyway nobody would PAY for a fake movie or art. The whole undertaking is utterly worthless in the monetary sense, if we ignore the marginal gain of few more subscribers to the chatgpt service which may use the feature for trivial personal entertainment. The usecase is OpenAI marketing, and that makes them look desperate, stupid and evil.
Sometimes I feel a bit hopeless when I see so many people enjoying AI slop while disregarding human artistry. Personally, from a game development background, I feel a bit fearful that big studios will chase the money and get rid of large parts of their teams, and that smaller studios will stop looking for passionate team members. In my view, the large scale adoption of AI seems like it will not only lead to pretty bad "art" being made - it will also lead to very lonely and atomized individuals, a core feature of an unhealthy society.
The question is, how do we, as artists, make people care about art? It seems like our current cultural incentives make it so that art is viewed more as content, as easily digestible and discardable noise to occupy our increasingly more distracted minds. For instance, I feel like there's often no time given to reflect on a show's impact on us - we just load up the next movie. Under that view, AI is a gold mine - infinitely repeatable and inoffensive content that does nothing new, interesting, or meaningful. It's all just bells and whistles.
Honestly, I'm not sure, but I think we definitely need to think about how we get more people on board. Surely some AI bros are already unreachable, given the mocking tone with which they regard humanity, but I'm certain there are lots of people who just aren't aware of how bad this stuff is. I don't know - what do you guys think?
Art can absolutely be licensed, that is the overwhelming majority of how artists are compensated for their work. I think the bigger problem is that if you adequately compensated every artist for using their art to build a model, plus paid NVIDIA for all the GPUs and the power company for all the power etc etc there would be no money left for the shareholders of OpenAI. They can't rip anyone else off so they rip off the artists. It's a time honored tradition in the music business at least, the artists get paid last, and often don't get paid at all.
The AI bois hate everyone that isn't a fucking computer nerd like them. They hate artists, writers, actors, anything that isn't code, because they can't understand it and and can't do it and feel that basic humanity is a worthless scam.
Art is truth, but tech CEOs art only resembles their truth - profit. Art is a gift freely given, thus degrades when profit is painted into the picture.
Well said.
We are clearly riding a wave in which hypocrisy is virtue, lies are truth, gross-outs are charming, and believing in human dignity or creativity is reviled as a sucker's bet.
What's especially sad is how this mentality invites losers to double down on their baseness.
Only someone who already felt like sad-sack would find a race to the bottom appealing. One day these people will wake up and find that their jokes are unfunny and everyone else decided long ago that follow-the-leader is a bore.
Maybe it won't be too late for them to get a life.
I fear the current political and social structures of the world only serve to valid that baseness as of now, but one can always hope....
Imagine promoting the fact that you stole Studio Ghibli's intellectual property. Altman is waaay too comfortable, but I hope it will be OpenAI's downfall.
In my experience as an artist I've found that most AI users do hold artists in contempt, I've had so many people be outright rude and patronizing when I point out how artists are being impacted by this tech. Their favourite word is "cope". They could just be a very loud minority (as is often the case with the Internet in general) but man are they LOUD. 😒
Mine too, and that's the assumption I made as well — always giving folks the benefit of the doubt. But they're at least both loud and powerful, if not particularly numerous; so a threat to be sure; and certainly dominant in the industry.
"Fair use" normally involves using *very small excerpts* of a work, for purpose of commentary on it. Not systematically digesting every pixel of the corpus.
If they put "a lot of thought" into this before doing, then they are even more of asses than I could believe.
The AI bros have almost a sociopathic level of disconnection from what people are saying about generative art, and what they are doing.
I feel nothing but sadness about this. I guess that makes me human still.
AI slop will continue to devalue art and, yes, we will keep moving onto the next shiny thing that distracts us. What we’re doing is destroying culture one prompt at a time. There is no ‘but’ that fixes this, no matter what the snake oil companies tell us.
AI should not be shortcut to human artistic creativity.
The resentment and disrespect of artists and their craft by these reactionary posters does nothing but vindicate people's already existing stance against AI. For all of their talk of "democratizing creativity", all they ever seem to do is punch down at creators who are already struggling during these hard times.
When did Altman change his X avatar? Was it immediately upon rollout of the new ChatGPT image generation features, or in response to the user-driven Ghibli trend? That timing matters. From the tone of your piece, it sounds like you lean toward the interpretation that Altman led and users followed—but does the evidence support that?
In testing the new image generator myself, I’ve found the safety filter to be surprisingly aggressive. Even seemingly innocuous prompts can trigger censorship for violating "community guidelines" or "safety standards." Applying the Studio Ghibli style appears to function as a workaround—it allows content through that would otherwise be blocked. I explore this in more detail here:
🔗 https://open.substack.com/pub/chatswithclaude/p/the-safety-layer-vs-the-social-primate?r=1ozb0&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
This is disgusting! So disrespectful to Miyazaki for him to do that
I guess it's the stories we tell that matter.
That of human talking to human.
When it's robot to robot... then who cares that matters.
I found this after reading the free part of another, positive review of this: https://www.thealgorithmicbridge.com/p/congratulations-on-the-studio-ghibli Then I followed the link to Open AI and tried my own:
"Draw a comic strip image in the style of Spirou and Fantasio where both characters sit in a red sports car, driving out of view into a scenery of open country. Chasing after them are several dogs. The car leaves a cloud of dust behind it. The sun starts setting and there are scattered clouds, some birds in the air."
Answer: "Here’s the comic-style illustration inspired by Spirou and Fantasio, with them driving a red sports car into the open countryside while being chased by dogs. Let me know if you’d like any adjustments!" Well, ChatGPT drew a Disney cartoon style image with Goofy-like characters in a car, where the car is chasing the dogs, not the other way around. I pointed that out to ChatGPT:
"Sorry, but this is American style 1950s not Belgian style 1960s comic strip. Also, it's the car chasing the dogs, not the other way around. Can you correct that?"
"Here’s the updated illustration in a Belgian comic strip style, now featuring Spirou and Fantasio in their red sports car chasing after the dogs. Let me know if you need further refinements!" This worked, the characters and the car looked for more Belgian-style. But out of five dogs, only two were chasing after the car, the other were in front or at the side.
Me: "Better. The dogs need to be chasing after the car, not going in front of it."
ChatGPT: "Here’s the revised illustration—now the dogs are correctly chasing after the car instead of running ahead of it. Let me know if you need any more tweaks!"
Me: "I'm sorry, but there are still two dogs going ahead, one after, and one beside, also two in the car. So of a total of six dogs, only one is actually chasing after the car. And the comic figure is again reverted to 1940 US comic strips. Did you forget my original instructions?"
ChatGPT: "It seems like I can’t generate any more images right now. Please try again later. Let me know if you'd like me to refine the description further so it's ready when the tool is available again!"
I'm really sorry, but ChatGPT should be called CheatGPT. Or dummy-dummy.
It's all just for show.
Parasitic behavior, the hypocrisy is disgusting, but... on the one hand I possibly couldn't eat as much as I want to throw up about the marketing. On the other hand, how is this new feature different from using text written by humans that did not give their consent expressly for LLM use and appropriation? Any LLM user, myself included, can't really throw the first stone.
Well, everybody pees into the swimming pool, but when you do it from the three-meter board more people tend to notice. We as societies seem to lack the appropriate decency or (legal) means to prevent freeloaders from succeeding. During this presidential term nothing will change. Let's wait for Musk's dressing down of "cry babies".
It's evil towards Ghibli, but what for? OpenAI or any their customer could not produce complete new movies "in the ghibli style" for profit, that would be quite an IP violation, and anyway nobody would PAY for a fake movie or art. The whole undertaking is utterly worthless in the monetary sense, if we ignore the marginal gain of few more subscribers to the chatgpt service which may use the feature for trivial personal entertainment. The usecase is OpenAI marketing, and that makes them look desperate, stupid and evil.
Sometimes I feel a bit hopeless when I see so many people enjoying AI slop while disregarding human artistry. Personally, from a game development background, I feel a bit fearful that big studios will chase the money and get rid of large parts of their teams, and that smaller studios will stop looking for passionate team members. In my view, the large scale adoption of AI seems like it will not only lead to pretty bad "art" being made - it will also lead to very lonely and atomized individuals, a core feature of an unhealthy society.
The question is, how do we, as artists, make people care about art? It seems like our current cultural incentives make it so that art is viewed more as content, as easily digestible and discardable noise to occupy our increasingly more distracted minds. For instance, I feel like there's often no time given to reflect on a show's impact on us - we just load up the next movie. Under that view, AI is a gold mine - infinitely repeatable and inoffensive content that does nothing new, interesting, or meaningful. It's all just bells and whistles.
Honestly, I'm not sure, but I think we definitely need to think about how we get more people on board. Surely some AI bros are already unreachable, given the mocking tone with which they regard humanity, but I'm certain there are lots of people who just aren't aware of how bad this stuff is. I don't know - what do you guys think?
Art can absolutely be licensed, that is the overwhelming majority of how artists are compensated for their work. I think the bigger problem is that if you adequately compensated every artist for using their art to build a model, plus paid NVIDIA for all the GPUs and the power company for all the power etc etc there would be no money left for the shareholders of OpenAI. They can't rip anyone else off so they rip off the artists. It's a time honored tradition in the music business at least, the artists get paid last, and often don't get paid at all.
The AI bois hate everyone that isn't a fucking computer nerd like them. They hate artists, writers, actors, anything that isn't code, because they can't understand it and and can't do it and feel that basic humanity is a worthless scam.