When AI becomes a paintbrush, is it art?

This week on 60 Minutes, correspondent Sharyn Alfonsi stepped into a new frontier of artistic expression: the rapidly evolving world of artificial intelligence–generated art. She profiled Refik Anadol, the 40-year-old Turkish American artist widely regarded as a pioneer of this emerging form.

Anadol doesn't mix acrylics or sculpt with stone. Instead, he paints with data.

For one recent work, he fed an artificial intelligence model 200 million photographs of Earth, drawing heavily from archives provided by NASA. The result is a sweeping, immersive digital installation — a living canvas of color and motion that feels at once cosmic and intimate.

"When I think about data as a pigment," Anadol told correspondent Sharyn Alfonsi, "I think it doesn't need to dry. It can move in any shape, in any form, any color, and texture."

It's a poetic description of a process rooted in code. His installations, projected across walls and ceilings, envelop viewers in constantly shifting landscapes generated by machine learning systems trained on vast image libraries. The effect can feel, as Alfonsi put it, "a little trippy."

"It is trippy," Anadol replied. "Because I think as artists we ask what is beyond reality."

The critics weigh in

Anadol's work has appeared in some of the world's most prestigious museums. But as A.I. art moves from tech labs to galleries, the art world is grappling with a bigger question: How do these creations stack up?

Jerry Saltz, the Pulitzer Prize–winning critic for New York Magazine, is both skeptical and curious.

"Right now, AI art seems to be an average of averages," Saltz told Alfonsi. Algorithms are trained on vast datasets of existing images, themselves products of countless influences. The result, he argues, risks becoming "vaster, and more average," rather than more profound.

For Saltz, great art emerges from something machines fundamentally lack: lived experience.

"I want the algorithm to experience death," he said. "I want the algorithm to know the feeling of feeling like you have a fat neck, or bad hair… I want to train the algorithm to experience carnality." Without sex and death, Saltz suggests, there is no art.

And yet, he doesn't dismiss the technology.

"I like to think of it as a material," Saltz said. "Artists use materials. A digital file is a material." To reject A.I. outright, he argued, would be like rejecting oil paint or the novel before engaging with them. "I wish it well. And I would never, ever ignore it."

Fear, replacement, and ethics

Part of the anxiety surrounding A.I. art is existential. Artists, like professionals in many industries, fear replacement, Saltz said.

"We all have a latent fear of being replaced by AI," Saltz acknowledged. "I guess I think that we will be on some level." His prescription isn't retreat — it's reinvention. Artists must become "better, or more useful, or more unique at what we do in order to keep our jobs."

The ethical questions are thornier. Is it fair — or legal — to train an algorithm on the work of other artists?

Saltz thinks so. Artists have always borrowed, referenced, and reinterpreted what came before them.

"There are no laws in art," Saltz said bluntly. "All art comes from other art."

Is it art?

Last year, artist Refik Anadol brought his vision to the Guggenheim Museum Bilbao in Spain. For that exhibition, he built a custom A.I. model trained on open-access photographs, sketches, and blueprints from the archive of Frank Gehry, the legendary architect who designed the museum itself.

The system processed Gehry's architectural legacy and reimagined it as a fluid, morphing, digital spectacle.

Saltz once dismissed a similar installation at New York's Museum of Modern Art as a "glorified lava lamp," dazzling but ultimately decorative.

Which raises the central question of this cultural moment: When a machine recombines humanity's visual history into something new, is that art? 

Photos & Video courtesy of Refik Anadol Studio, Guggenheim Museum Bilbao, Tom Ross & Getty Images.

The video above was edited by Scott Rosann.