- Generative AI can create images from written prompts, but the art world has had mixed reactions to it, according to artists and gallerists interviewed by CNBC.
- While some are concerned about the impact on intellectual property and creativity, some welcome the potential of generative AI.
- Search engine Have I Been Trained allows people to see if their work or images are being used to train some of the larger language models behind generative AI.
Artist Refik Anadol uses generative AI to create images. This can be seen as part of the 'Echoes of the Earth: Living Archive' exhibition at the Serpentine North in London.
Hugo Glending | Courtesy of Refik Anadol Studio and Serpentine.
Like many industries, the art world is grappling with how to best leverage artificial intelligence, especially its newest form, generative AI.
Image generators such as Midjourney and OpenAI's DALL-E 3 can generate images from written prompts, and such technology can be used to create magazine covers, win art awards, and even appear on the pope. It is used to dress white down jackets.
Some artists CNBC spoke to described the technology's potential as frightening or threatening, or expressed concerns about copyright. But they also said they were excited about what generative AI could bring.
Installation artist Rubem Robierb told CNBC by phone that he was “shocked” when he first saw what generative AI could do. “In its early stages, [generative] AI can create more images in less than a second [than] The human brain can even process it. “This is not necessarily a good thing, but we are all being forced to experiment with it here,” he said in a follow-up email.
Robierb specializes in sculpture, and in addition to a piece titled “Dandara” that was exhibited in New York City in memory of Dandara dos Santos, a transgender woman murdered in Fortaleza, Brazil, Robierb also created a large sculpture called “Dream Machine.” Produced. Butterfly wings commissioned by Celebrity Cruises for their multi-billion dollar cruise ship Edge.
Artist Rubem Robierb and his sculpture “Dream Machine”. Robiab wants “legal boundaries” to be introduced to protect artists' intellectual property.
Rubem Robierb
The artist, who is based between New York and Miami, said he has not used AI in his work yet. But he said doing so was “not a matter of choice,” adding that he was considering when and how to use it.
“We can also see it as a threat to creativity, as it currently exists. [generative] AI retrieves information from known images, known artwork, and known artists to complete tasks. Legal boundaries need to be put in place to protect intellectual property,” Robiab said.
In Europe, the European Commission's AI law aims to regulate technologies according to how risky they are deemed to be in terms of public rights and safety, and according to a December press release, around It is said that it is likely to be implemented within two years. .
Using generative AI in an ethical manner is a key consideration for London gallery Serpentine, which has been developing AI projects with artists since 2014, says CEO Bettina Korek.
One of the gallery's current exhibitions, “Echoes of the Earth: Living Archive” by Refik Anadol, uses approximately 135 million “publicly available” coral images to Large-scale AI-generated artworks are on display, including the creation “Artificial Realities: Coral.” accessible online,” according to a press release.
“AI seems far removed from the experience of humans like us, but Refik has created such an immersive and sensory experience,” Kolek told CNBC in a video call. . “The audience is really encountering art first and technology second,” she says, adding that Anadolu uses “ethically sourced” data to train the AI that generates the images. He added that he focused on the importance of
Artist Refik Anadol used generative AI to create works of art. This can be seen in the 'Echoes of the Earth: Living Archive' exhibition at Serpentine His North Gallery in London, UK.
Hugo Glending | Courtesy of Refik Anadol Studio and Serpentine.
Anadolu used what he called the “Mother Nature Model.” Data from the likes of London's Natural History Museum and the Smithsonian Institution was used to train his AI to create content for the piece, titled “Living Archives: Models of Nature.” It was first unveiled at the World Economic Forum in Davos, Switzerland, in January.
Ethically sourcing data to train AI is part of “a larger conversation that we have with artists,” Kolek said, adding that the fourth issue of Serpentine, published in March, The Next Future Art Ecosystem report calls for public institutions to “understand themselves as intermediaries in the role of AI in society.”
Other galleries are also embracing AI-generated art, such as 37xDubai in the United Arab Emirates. Danilo S. Carlucci, the gallery's founder and CEO, said the exhibition “Generative: Art & Systems” will feature artists such as Julien Españon who combine design, code and art. The artist's works will be exhibited.
What are we doing to replace the human experience?
Shane Gaffog
artist
When asked if generative art can match the value of human-made art, Carlucci said in an email to CNBC that generative art requires creativity and skill. “Some of the artists participating in our exhibition are highly skilled and understand code very well. “Like the art, the stories behind their pieces have thoughtful…messages,” he said.
At the Serpentine, the gallery's art technology team is working on a number of AI projects, including “What does it mean to be an artist in the age of AI?” by artists and musicians Holly Herndon and Matt Dryhurst. It also includes an exhibition that explores “Dark Corridors.'' According to a press release, it is scheduled to open in the fall.
Dryhurst and Herndon are also co-founders of Spawning, an organization focused on data governance for AI. One of the company's products, the search engine Have I Been Trained, allows people to use their creations and images to train some of the larger language models behind the AI they generate. You can see if any of them exist and have the option to prevent them from being used in the future.
AI platforms Stability and Hugging Face are two of the generative platforms that use the Have I Been Trained registry, and Spawning is “actively courting” both OpenAI and Midjourney, said co-founder and CEO Jordan Meyer said in an email to CNBC.
Abstract artist Shane Gaffog has mixed feelings about AI. In a video call with CNBC, he described AI as a “tool.” “Some parts are scary. Some parts are exciting, because it allowed me to release what I was hearing 'intuitively',” he said.
Gaffog has synesthesia (a sensory condition that equates individual colors with specific musical notes) and wanted to create a musical work based on his art that could be performed by a pianist.
He asked software developers to help him make it happen in a piece called “Sounds of Color,” which is part of an exhibition opening in Venice, Italy, starting April 20, but some development realized that people want to replace the human element with technology.
“They wanted it to be completely AI-generated based on my movements, not my drawings. And…the human element is removed. And I just said, 'No, that's not what I meant.' 'No,''' Mr Gaffog said.
California-based artist Shane Gaffog collaborated with an AI software programmer and pianist to create an exhibition in Venice, Italy from April to November. He said AI technology is both “scary” and “exciting.”
Shane Gaffog
Another developer wanted to create a hologram of Gaffog that could create new art. “He said… once we record every move you make, we'll be able to forever generate a new picture long after you're gone,” Gaffog said. He also declined the proposal.
Gaffog, who created “The Sound of Color” with AI software programmer Jonah Lynch and pianist Anthony Cardella, said he cried when he “heard” his painting played for the first time. He said he washed it away. “I heard everything [musical] “I was influenced by what I listened to while I was painting,” he said.
Gaffog said he hasn't experimented with generative AI programs, but that people have shown him images created that way. Creating your own art is “the joy of discovery,” he said. “What are we doing, displacing the human experience? … Hopefully … it will wear itself out and it will no longer be a brave new world,” he says in the context of art. He talked about generative AI.
Robierb had similar sentiments. ”[An] Original artwork is only original if it comes from a person. There's nothing like that, original creativity.I think someday we should go to art fairs and label our works [that are] It's man-made,” he told CNBC.