Why Artificial Intelligence Is Making Internet Art – And Art – Less Original

The “The death of art” It’s a story we’ve heard many times: replace “video” and “radio” with the common adage and you get “AI killed the artist’s star”. But there is more to this current iteration of concern about art than those that preceded it. It comes down to the fact that it’s not the medium that has changed here – it’s the process itself. Artificial intelligence software allows art to be created in seconds, replacing labor and a skill that takes years to acquire.

But with AI art, there is a more fundamental problem – one that begs the question of the purpose of art itself.

Certainly, AI art captures a sense of wonder and awe, capturing the curiosity that resides in our souls and raising the mysteries of a cosmic order that we are simply too young to understand. For a taste, take this series of AI photo Imagine what public figures who died prematurely would look like today, in their old age. Or take the “Théâtre D’opéra Spatial” – a piece that won First prize At the Colorado State Fair. It depicts a group of people staring into a bright void, stylized like a Renaissance painting. All it took to create it was a few text messages fed to Midjourney, an artificial intelligence program. One Twitter user pointed out similarities in style with French Orientalism in art that evokes the exotic. “She uses details to dazzle us without meaning… In the style of the Orientalists, they often mixed and combined many different styles of Levantine and Egyptian architecture and culture without regard to what they meant or where they came from, just like how AI art re-integrates meaning without its context.” pointed.

others optimistic The emergence of the art form – if it can be called – could herald another revolution in art itself.

“When photography appeared, so did Impressionism. What will appear now, when every conceivable combination of color and shape can be configured in just a few seconds by an algorithm?” Requested One Reddit user.

But with programs written in the Global North and datasets containing their corresponding bias, there is a bigger question to be asked about whose gaze is represented in the art of AI, and what that says about the model itself.

The art of artificial intelligence has started conversations about who an artist is, and what we also consider art. But beyond that, there is a more pressing question regarding the costs of the AI ​​programs that generate art. What happens when a machine takes seconds to create something – also based on the myriad of existing human-made art? Here is where the ethics get murky, and the cultural conversation about the art of AI quickly becomes philosophical.

The thing about AI art is that it’s easy to make. There is a vague definition of Who are its “artists” Given that while the compositional intent comes from a human being, the actual art comes from a machine. This, too, is a machine that gathers information from a trove of pre-existing data sets — itself awash in bias. Previous research has shown that artificial intelligence It reinforces racist and sexist stereotypes Because it does not identify biases in the datasets that are fed.

“They don’t appear to have done any regulation of the data sets used… and then they ask the user to be careful not to promote harmful stereotypes – effectively washing their hands of any responsibility for this,” notes Divyansha Sehgal, a researcher at the Center for Internet and Society. “All technologies are political and all design choices have values ​​embedded in them,” she adds.

Plus, Anyone can use open source AI Technical software for making (mostly) anything – whatever they can’t get past. If earlier people didn’t go through the trouble of creating inappropriate or obnoxious art with their own hands, now they can do it simply by typing a few prompts into the program – and all it takes is a few seconds. This, along with the fact that it can be taken advantage of for profit to the few people who control the program, is by definition, the antithesis of art as a public good. “Technology is increasingly being deployed to make temporary jobs and to make billionaires richer, and not many of it seem to benefit the common good enough… AI art is part of that. For developers and tech-minded people, that’s great, but for illustrators, it’s It is very annoying because you seem to have eliminated the need to hire a painter,” He said Cartoonist Matt Boerse, responding to The Atlantic using an AI-generated image in a newsletter instead of hiring and paying a human illustrator.

Take the latest development that allows users Modify Real human faces using DALL-E 2, an AI art platform stylized after the famous surrealist artist Salvador Dali. Filters and photo-editing technologies have already raised questions about the authenticity of art: AI may not only intensify this debate, but extend the scope of the damage infinitely.

Related to swaddle:

The world is turning to artificial intelligence weapons to reduce the losses of war. What could be wrong?

The other problem with AI art is that it is relatively easy to produce. That’s why artists are concerned that they are the first frontiers for robots to replace their jobs – a prospect that should concern us all. “I worry about compromising audience and market expectations. Just because the art of AI is available so cheaply, it may flood the market by putting out ‘content’ that our social media algorithms love so much,” Segal adds.

Moreover, by borrowing from already existing images and ideas, AI art may respond to original claims but not necessarily be creative about them. In the process, it can be infringing on existing copyrighted work – and more importantly, masking something much more sinister – copyright infringement, mass production, copying, and ultimately content creation – as creativity.

“…the question that arises is whether this is ‘creation’ or just a creative emergence that hides unethical artistic and artistic practices,” Deepa Singh, a researcher in artificial intelligence ethics in the University of Delhi’s Department of Philosophy, tells The Swaddle.

While some have argued that AI art is more accessible and deals with the gatekeeping inherent in the art world, Singh and other artists themselves refute this claim. “The basic idea behind AI art generation is not to have access to art per se, but to make machines do the art and see where it goes from there,” she says. The problem with this lies in the underlying logic of AI as a tool in and of itself, which makes the approach flawed: “The AI ​​value system is the value system of Silicon Valley, which in turn is also the (mostly) white male value system. Venture money who turned into technical missionaries and technical utopians.”

The ease of creating art – combined with copyright and existing data set issues – leads to a circular situation, in which the AI ​​artist himself is now Struggling to claim his own art. One of the internet’s most famous artists, Greg Rutkowski, is said to be now more sought after than Picasso in software claims. But now, with his own style permeating cyberspace while crowding out his own original work, Rutkowsi hardly has any claim to his own style. “It’s only been a month. How about a year? I probably won’t be able to find my work there because of [the internet] He told the MIT Technology Review.

Related to swaddle:

AI rappers forced out racial stereotypes, showing how technology is commodifying culture

Content creation is not an art form. For a better analogy of what we’re missing out on in the art of AI, turn to AI Scripts – themselves an online joke. Feed a few prompts to a text box, and the AI ​​software can write a novel, text, story, essay or term paper in seconds. Is it good, and does it make sense? can not say. They are just trying to make different changes to what is already there. Synthetic art lacks its own intrinsic psychological meaning for the subject. AI agents do not make art; Rather, they are repeating art.” Writes s. Will Chambers, in a newsletter.

When something is easily reproduced, reused and standardized, it loses something of its essence. At least, that’s what the cultural theorist Walter Benjamin said – almost a century before the advent of the art of artificial intelligence. He was talking about the ease of reproducing art in the industrial age – where mechanical reproduction denies art from its “aura”. This is the art of artificial intelligence – a compelling endless loop of images all lacking the aura of how they are different iterations of the same few things. It can be argued that nothing new, challenging, or disruptive is possible when the means of creating art are controlled by big technology.

Given how artists borrow themselves from tradition, experiment with orchestration, and play with ideas, is the machine making things worse? It can be said, yes. It takes away from the creation process – adding to the endless content culture, proliferation, and speed that has come to define consumer culture. We may no longer have to wait to see what happens to art when it reaches its maximum consumption – we may already be here.

Leave a Comment