Did an AI Paint That?

“My kid could paint that” is a frequent joinder to art that viewers struggle to understand. But “my computer could paint that” may become the challenge to art that has secured its status as a masterpiece.

DALL-E 2(Opens in a new window), an OpenAI-created system that translates images into art, has taken over social media, emphasis on “media.” The AI ​​and its unrelated copycat, craiyon(Opens in a new window)have been used to create such works as “an ancient Egyptian mural of an Egyptian animal god using a computer” and “a medieval painting of people wearing VR headsets.”

DALL-E 2 has been available to a select few who have moved up from its waitlist, and craiyon is open to the public. The work that has been produced via these programs has caught the attention of many, including Tidio, which decided to pose some questions(Opens in a new window) about it to 842 people on art, AI, and technology subreddits (scroll down for the full infographic).

Tidio showed several images to his audience and asked of each: Did AI make this or did an artist? A surrealist painting from the 80s was judged as painted by a human by 68% of people surveyed—while a DALL E 2 creation was deemed human-made by 73%. Nine percent of those surveyed said it was easy for them to tell whether AI or a human was behind an artwork, but Tidio pointed out that not all of those people were correct in their assessments of the art they were shown.

But since beauty is in the eye of the beholder, we should note that the artist-versus-AI debate doesn’t matter to everyone. Almost 67% said AI-generated images are still a form of art. The positives they associated with AI-generated art are an easy conceptualization of ideas (25%), a good source of inspiration (24%), a new means of self-expression (17%), the fast creation of images and cartoons ( 16%), and a better understanding of our minds (8%).

But some fear that DALL-E 2 can fool people into thinking that they’ve seen photographic evidence of momentous events in history—like, say, Super Mario getting his citizenship at Ellis Island. Deepfakes and manipulation (41%) were at the top of the list of worries about what text-to-image generators can wreak. They were followed by misgivings that AI models perpetuate biases (27%), artists could lose jobs (18%), and AI could create pornographic content (10%).

Tidio said that some surveyed pointed out that AI had not created their digital canvases out of whole cloth; the images rely on existing artworks, and those artists are going uncredited and uncompensated. But perhaps stealing from other artists is the most human thing AI can do.

What’s New Now to get our top stories delivered to your inbox every morning.”,”first_published_at”:”2021-09-30T21:30:40.000000Z”,”published_at”:”2022-03-29T17:10:02.000000Z”,”last_published_at”:”2022-03-29T17:09:22.000000Z”,”created_at”:null,”updated_at”:”2022-03-29T17:10:02.000000Z”})” x-show=”showEmailSignUp()” class=”rounded bg-gray-lightest text-center md:px-32 md:py-8 p-4 mt-8 container-xs”>

Get Our Best Stories!

Sign up for What’s New Now to get our top stories delivered to your inbox every morning.

This newsletter may contain advertising, deals, or affiliate links. Subscribing to a newsletter indicates your consent to our Terms of Use and Privacy Policy. You may unsubscribe from the newsletters at any time.

Leave a Comment

Your email address will not be published.