Glibhi-fication Is AI at Its Worst

Art without intention

Look at me! I’m now some sort of Glibhi-style animation expert because I can prompt an AI chatbot, even though I only learned about its existence last week and didn’t spend one second thinking about this. (Shit, it doesn’t even resemble me, damn AI!)

I don't get art.

It's not for lack of trying. I once visited a wall of banana skins pinned in different positions, all in various states of decomposition. One exhibition I went to was an empty room, aside a toilet in the middle that had someone's signature on it. I've seen a display that just had empty bottles placed around the floor, seemingly at random. At my final year degree show, there was a woman whose entire project was a fuck you to the patriarchy (or something), and her exhibit was this little boxy room with a screen inside that played a video of her masturbating. I know this is true because I was an immature 20-year-old who obviously wanted to check out the serious art on display. The pièce de résistance? Her parents apparently had no idea when they turned up to see what their daughter had been working on.

I've stared at dozens of apparently famous paintings, objects and displays and been left utterly confused by what the creator is trying to convey. No matter what people write on the little card on the wall, or whatever they title it, or whatever story they create to explain the concept — my brain rarely connects the dots.

Again, I don't get art.

But, what I do get is that these people are passionate about their ideas, beliefs and opinions, and have a burning desire to express them in whatever way they see fit. To them, it means everything. Even if it's through the medium of rotting banana skins or playing with one's bits, they do it with heart and soul, painstakingly translating what was in their mind into something unique for the world to see. If you spend time in an art museum and you will see things you will never see the likes of again.

That's the beauty of human creativity.

What I'm getting at here is that—whether you or I understand it or not—behind most art and other creative outputs like writing, music, photography, etc., lies intention. I say most because, of course, like everything else, the creative world is full of cheap imitations, knockoffs, repetitive themes and styles, and money grabs. But that's a subject for another day.

No, that actually leads perfectly to the recent updates in AI-generated images.

On the face of it, the gap has been well and truly closed. What is real and what is generated? Fucked if I know. OpenAI's latest update allows image generation through the chat interface, which means prompts can be more complex and, in theory, offer better results. The initial flood of output across socials shows that this update is a significant improvement. Those bullish on the industry are rejoicing that the technology is moving closer towards completely automating creative output, which is a good thing to them I guess? What I would say is if you're a run-of-the-mill marketer right now who is pumping out generic visuals, you are in trouble.

But watching the great Glibhi-fication that played out across social platforms over the last week brought back the million-dollar question, one that generative AI as an industry hasn't been able to answer yet.

What is the point of all of this?

Is the future of this all just art, because we can?

The sheer speed and scale at which pictures were turned into Studio Ghibli replications drove home how devalued any form of creativity is becoming. Hayao Miyazaki, the original creator, is someone who has fought hard to retain the identity of his films. He's also quoted as saying AI tools are "an insult to life itself." He would have been last in line to let AI's train on his work — something that has clearly happened — and in a way, the torrent of imitations feels like a big middle finger to him and his work, making a farce out of decades of artistry. Thousands upon thousands of images, from people who don't care for Glibhi, or what it means, or what it represents, or what dedication and craft was required to create it, churning out prompt after prompt for the sake of it, all to be part of a collective meme moment. (Yes, the feature image of this article is to demonstrate this).

Put another way, is generative AI going to exist solely to waste unfathomable amounts of resources and compute producing creative output that we don't care about, or can't be bothered to spend more than two minutes learning how to create? Burning billions of dollars to produce pointless, meaningless and intention-less content at relentless volume, all for here today, gone in 20-minute braindead trends on social media platforms, like turning pivotal moments of history into a cartoon or generating random marketing posters of product images?

I love the way

Erik Hoel

frames it:

The semantic apocalypse heralded by AI is a kind of semantic satiation at a cultural level. For imitation, which is what these models ultimately do best, is a form of repetition. Repetition at a mass scale. Ghibli. Ghibli. Ghibli. Repetition close enough in concept space. Ghibli. Ghibli. Doesn’t have to be a perfect copy to trigger the effect. Ghebli. Ghebli. Ghebli. Ghibli. Ghebli. Ghibli. And so art—all of it, I mean, the entire human artistic endeavor—becomes a thing satiated, stripped of meaning, pure syntax.

We're almost three years into this AI mess, and the most popular use case still seems to be, "What if I made this popular thing look like another popular thing, but without the effort?" 

Maybe this is the point of it, at least right now. Art without effort. Art without thought. Art without consideration. Allowing people to create work not to demonstrate passion, joy or skill-building, but to jump on a "hey, everyone else is doing this thing" bandwagon? A soft-entry point into a reality where AI steals your livelihood.

The last week has shown us that we are not far away from the tipping point where AI and human-made output become indistinguishable from each other. AI tools will soon allow people to create work that replicates the most famous creators. They will soon allow people to produce hyperrealistic marketing ads. They will soon allow people to do creative writing, music and video production to a level on par with, and then likely surpassing, that of human beings. 

I'm going to end on a point of hope here (*reader gasps). What we've also seen is that, even when people can create almost anything they can dream of, most people will use technology like sheep following the herd, doing the things other people are doing just because they can, not because they want to. 

To make great things, things that transcend moments and become part of history, that define movements, that create cult followings, or that genuinely move people, you need to value the medium and have intention. 

When anyone can create anything, and the barrier to entry — hell, even mastery — becomes so low, all we'll have left is the intention. And in that little space, there is still room for humans to produce great things that mean something.