Explainer: Meet the AI ​​art generator that’s breaking the internet

Explainer: Meet the AI ​​art generator that’s breaking the internet

Ever wonder what MP Kristyn Wong-Tam would look like next to a capybara? What about what Doug Ford might look like if he appeared in the 2019 horror film “Midsommar?”

A hilarious internet tool can show you … kind of. It’s still learning.

A new AI model which generates artwork from human-inputted text is garnering attention across the internet and filling up our social media feeds.

Users are prompted to type in a phrase describing their desired art, and a grid of nine images is offered up for your consideration.

MPP Kristyn Wong-Tam befriends a capybara

According to a paper by Dall-E creator Boris Dayma, the model learns “by looking at millions of images from the internet with their associated captions.” The model retains some memory — individual images and search terms — but it continues to learn when faced with unique inputs like, for instance, “Doug Ford drinking bubble tea.”

Over time, the AI ​​model will continue to improve, creating more accurate images from text-based prompts.

The generator’s name is a reference to the surrealist artist Salvador Dali and cartoon robot WALL-E.

Who did we miss? Click here to send us your best AI art.

Here are some of our favorite images we’ve created using this model at the Star:

John Tory at Coachella.
Kyle Lowry in
Prime Minister Justin Trudeau with Willy Wonka.
The Weeknd in Candyland.
Carly Rae Jepsen on a TTC streetcar.
Toronto as
Raptors Superfan Nav Bhatia in
Shrek with a Tim Horton's Iced Capp.

JOIN THE CONVERSATION

Conversations are opinions of our readers and are subject to the Code of Conduct. The Star does not endorse these opinions.

.

Leave a Comment

Your email address will not be published.