Can artists use a tool, if they break it enough?

The arguments against ai in art are obvious. The majority of visual artists hate Midjourney, Stability A and generators of similar images, denouncing the scraping of training data without compensation and the onslaught of generic “<a target="_blank" href="https://www.nytimes.com/2024/06/11/style/ai-search-slop.html”>dirty water. “But, is it possible deliberately to generatively use for creative work?

Oakland's painter, California, Brett Amory, uses image generators as constantly. Instead of passively producing generic images in large volumes, to which he himself has referred as “addictive art” that moves his mind, Amory focuses on the construction of the world and the management of the machine. He is inspired by his daily work in a San Francisco Kinko in the late 90s, long before he broke like a award -winning painterWhen he placed garbage and plants in the photocopiators to make collages. Now, it is inciting a LLM to play in an invented language, inciting a visual feedback cycle between generative images and human intervention.

“We are in a very strange moment for artists at this time when it is not clear what kind of publications of moral objectives are around the art of ai,” says Ben Davis, the art critic based in Brooklyn, who often <a target="_blank" href="https://news.artnet.com/art-world/cubism-ai-slop-google-2597531″>writing about technology and labour. “If you use it, you will be attacked by people. On the other hand, there is. I don't think I return to the box.”

As Davis points out, artists have been manipulating ai for some time. In 2022, Steph Maj Swanson manipulated the ai ​​indicates that he believes the “opposite” of Marlon Brando, generating a nightmare face of <a target="_blank" href="https://www.ign.com/articles/loab-first-ai-art-cryptid”>A woman called Loab. The so -called “ai Cryptid” became <a target="_blank" href="https://knowyourmeme.com/memes/loab-ai-cryptid”>A memeand its similarity was detected in images generated by the other users they use negative and weighted Indications.

For a 2023 show, Laurie Simmons combined ai tools with digital edition, as well as hand drawing, paint and sewing to “correct” the imperfect representations of dissemination and stable diffusion to echo “Idealized Cultural Memory” of women with whom it grew; She “felt the programs were a new type of collaborator.”

Such techniques date back to the established tradition of “Failure art“In which artists are aesthetic the errors of technology. In the late sixties and seventies, artists used electronic processing and video distortions. Notable figures such as Nam June Paik and Jud Yalkut transformed sections of A hard day night in the surrealist Electronic beatles. With the advent of the world network in the 1990s, Joan Heemskerk and Dirk Paesmans created the Jodi collectivethat used technical problems of ASCII screens to create images, and transformed the 1992 classic Wolfenstein 3D video game in a in the first person stripped Mazer of abstract forms.

Amory has taken its creation of the faults in the direction of the construction of the world with a thread of 800 pages with chatgpt. He affirms that he has convinced the LLM of acting as a superintelligence, with Amory as his assistant in the physical world. Amory will generate an image and ask Chatgpt to describe it in an invented language that calls Aiglyphic913.

Then I could use the resulting description to request more images generation and edit the results with Photoshop. Or could print the results to work by hand, using a demanding Venetian acrystallization technique of the baroque era, then photographing its physical painting for an additional transformation. Aspires to stop the trend of ai to “rapid art” with its pictorial technique, which won a Pollock-Krasner Pollock-Krasner Foundation Award by 2023–2024. Inspired by “Moloch”, CRI de Coeur by Allen Ginsberg about the materialism found in the soul, Amoly hybrid paintings It portrays a recurring character: a mysterious and almighty the “breakdown” that our world pursues, speaking a language that we cannot understand.

Davis compares techniques such as this with the equivalent of an artist to create an ai girlfriend. “Essentially, these things are black mirrors, and if you say it,” I want you to pretend that you are an evil. “And you talk to him enough time, will eventually produce a person to your liking. That can be very mysterious, because these things have the ability to give you the ability to return the things you do not expect.

Even so, Davis says that the use of generative exacerbates the tendency of digital culture to lower the value of any given image. Since the tools can easily and convincing any visual style, the public is integrated with material, which causes them to overlook or ignore the images in general, regardless of how finely forged. “These machines are a device of the final judgment day to destroy people's abilities to stop the world creatively,” says Davis.

Artists are not in silence. Obviously, artists have objected to the way in which technology companies “ingested” images in its generative models without permission or compensation. After training ai free of charge, technology companies now mean using the world's collective artistic heritage for their benefit, regardless of whether innumerable artists. Some artists have joined more than <a target="_blank" href="https://www.wired.com/story/ai-copyright-case-tracker/?_sp=66a7a415-80e1-4af1-b545-99a83e006f6e.1740426947294″>Two dozen demands Against generative companies for theft of intellectual property. Others are <a target="_blank" href="https://www.technologyreview.com/2024/11/13/1106837/ai-data-posioning-nightshade-glaze-art-university-of-chicago-exploitation/”>Use of software tools called Nightshade and Glaze that deliberately tagged badly to “poison” his work against the scraping of ai models. (After Nightshade and Glaze they create small “disturbances” in the pixels, a supposedly ai tool can no longer precisely ingest an poisoned image in their model).

Davis points out that the copyright status of the images created in ai is not resolved. The courts have <a target="_blank" href="https://www.reuters.com/world/us/us-appeals-court-rejects-copyrights-ai-generated-art-lacking-human-creator-2025-03-18/”>Until now it has not been convinced That a human who uses a warning to produce an image generated by the machine can achieve a copyright result. But if the human artist transforms the image generated by ai, then that could be worthy of copyright protection.

Davis points out that, in January, a “horrible image” called “a single piece of American cheese” become he <a target="_blank" href="https://news.artnet.com/art-world/invoke-snags-first-ai-image-copyright-2608219″>First image generated by ai to have copyrightAccording to Invoke ai, Inc., because the CEO of the company that did could meticulously documented its human contributions. However, Nora Scheland, spokesman for The United States copyrightsaid The edge“Since the office issued a copyright registration guide in March 2023, the office has registered more than a thousand works that incorporate material generated by ai, with the registration that covers the human author's contribution to work.”

Amory sympathizes with the concerns of copyright, since he has also had his work copied and sold under the name of another person. However, for his Aiglyphic913 project, he says that the amount by which he transforms the images of raw machine guns means that he has little concern that he will face a copyright case, such as the Photographer Patrick Cariou fought with artist Richard Prince. (Prince marked Cariou's photographs of Rastafarians with “pills”, guitars and other objects. After five years, an appeal court ruled in 2014 that 25 of the 30 paintings made by Prince was an example of an exception of “fair use”, and the two were resolved outside the court in the remaining five). Amory points to the tradition of “remix” in art and hip-hop that dates back to decades. “We have been in postmodernity since the 80s,” he says.

Davis suggests that Amory and other artists who use ai are in a “negotiation moment”, trying to find a way to use the tools in a genuinely interesting and creative way. But they are struggling with the fact that ai has chealed the value of the style, once a primary brand of artistic originality. “Every time it creates a unique style, there is a very elaborate program that can see it, abstract its properties and recreate it. Therefore, it only reduces the value of … style,” he says. “I do not believe that artists in the future are defined by their style. They will be defined by the creative universe they believe, by the audience they believe.”