Davis Institute for AI Issues a Call for Art

Art made with the help of artificial intelligence can be controversial, but it’s also a way to showcase human creativity

This image was made by the Davis Institute for Artificial Intelligence by entering the prompt “London Street rainy night expressionism” into the program Dream by Wombo.
By Abigail Curtis AI-generated art contributed by the Davis Institute for Artificial Intelligence
October 13, 2022

It all began with a hunt for a fun new mascot for the Davis Institute for Artificial Intelligence, the first cross-disciplinary institute for AI at a liberal arts college. 

When Amy Poulin, the institute’s assistant director, typed “blue mule with a computer” into an online program that uses AI to turn word prompts into images, she was captivated by both the process and the pictures created, such as a cartoonish mule holding a big computer.

Her enthusiasm caught the attention of Amanda Stent, the director of the institute.

“Blue Mule with Computer” by Dall-E mini

“Amy said to me one day ‘that AI is so creative.’” Stent recalled. “And I said, ‘Let’s have a conversation about that sentence.’” 

That ongoing conversation, and the national interest in—and sometimes concern for—the new field of AI-generated art, inspired Stent and Poulin to hold Colby’s first-ever AI-generated art exhibition

Students, staff, and faculty are all invited to participate by making art based on the theme of “The Lived Environment” (the Davis AI 2022-23 theme). All entries, including the prompts used in their creation, must be submitted by end of the day Oct. 19. Ten will be chosen to be displayed as part of an exhibition that opens Nov. 3 in the Olin Study and Collaborative Space.

“We want to start this conversation because next year at Colby is the year of the arts,” Stent said. “We decided to have an art exhibit to focus on how AI can facilitate human creativity, not replace it.” 

AI-generated art is well worth thinking and talking about, she said. New programs like DALL-E 2, Midjourney, and Stable Diffusion are making AI art accessible to more and more people, including those with no computer-coding experience. 

The growing popularity of these tools has led to an increase in concern and controversy, too. In a recent example, one of the winners in this summer’s annual art competition at the Colorado State Fair used Midjourney to create his piece, which won the top prize for emerging digital artists. But the prize led to a fierce backlash from artists who essentially accused him of cheating, according to the New York Times

The debate about ethics is important, Stent and Poulin said. But for those who worry that AI means the computers (or robots) are in charge, it’s good to remember that human creativity is the engine that drives machine learning and artificial intelligence. 

“London street rainy night Impressionism” by Dall-E

“It’s the programming and the people who do the programming who are the creative ones,” Poulin said. 

But AI-generated art does pose some clear dilemmas, too, such as how it can be used to create so-called “deep fakes.” Such created videos or photos digitally replace the likeness of one person with another, typically for malicious purposes. 

“Deep fake and other risks are very clear,” Stent said.

So far, interest in the art show has been high with several artists already submitting entries and a Colby professor emeritus reaching out to ask if someone from the institute could talk to their senior citizen’s art class about AI-generated art. 

“People are excited,” Poulin said. “This art allows people who have no ability in drawing to have something they can express their creativity with. And you can take these pictures and edit them and change colors and put your own creative stamp on them. You’re only limited by your imagination.”