Google Enlisted Artists To Reimagine ‘Alice In Wonderland’ Using AI
Google asked four visual artists to reimagined original Alice in Wonderland illustrations using AI, and to quote Alice herself, things are getting curiouser and curiouser.
CG artist and creative director Haruko Hayakawa renders Alice as a porcelain figurine of a young Asian girl in pigtails, a black sweatshirt and sweatpants and red Mary Janes, and the white rabbit as a Japanese folk-style toy on wheels. Graphic designer and art director Erik Carter presents Alice as redheaded with green eyes and yellow skin, and the white rabbit as a chipper bunny in a red bowtie. Hayakawa’s scenes have a surrealist pop art vibe, while Carter’s pixelated images look like something out of an 8-bit video game.
“What makes AI engaging is that it is unpredictable,” Carter, an adjunct professor at New York’s School of Visual Arts, said in an artist’s statement. “You can only pre-plan so much, but it will still take these left turns you can’t account for.”
For the project, titled Infinite Wonderland, the artists used a small set of their own original images to train Google DeepMind’s Imagen 2 image-generating AI model in their style, making creative tweaks as needed until the model’s outputs authentically reflected their own. They then prompted the model to illustrate Lewis Carroll’s 1865 children’s classic through their eyes.
“The role artists play in crafting these AI models is vital and impacts the whole creative process,” Hayakawa said in a statement.
Google Labs, which experiments with AI technologies, says it initiated the project as part of its ongoing effort to involve artists in shaping AI tools. An interactive online experience lets viewers read the timeless tale with accompanying AI-generated images in each artist’s style or see their four distinctive visual interpretations of the same passages side by side, along with the prompts used to generate each image.
It’s also possible to go old school and read the text alongside pencil sketches by John Tenniel, whose original illustrations have informed countless interpretations that followed.
Generative AI has, of course, elicited a vast array of passionate responses from artists. Some fear the tools will steal their work to train datasets, cost them their jobs or alter the very nature of the creative process. Others are excited about AI’s potential to steer them in wild and wondrous directions.
“I am always interested in trying different techniques, whether that’s glass or drawing or fabric,” Shawna X, another artist who participated in the Infinite Wonderland project, said in an artist’s statement. “I think AI is just another tool, another technique, that will help me create.”
Shawna X’s surreal, airbrushed Alice in Wonderland world explodes in vibrant hues. Colorful spots cover the bright purple body of her smiling, hookah-smoking caterpillar, and horns decorated with green and blue patterns sprout from its head. The artist’s version of Alice, modeled after her mom, has gray glazed skin, red lips and an iridescent blue-black bob.
Art director and technologist Eric Hu, on the other hand, imagines Alice as a young Black girl with flowing blond locks, blue eyebrows and modern garb, and the caterpillar as giant and pink. In the hands of Hu, Nike’s former global design director for sportswear, whimsical wildflowers populate the landscape of the lush, cartooney world Alice discovers when she tumbles down the rabbit hole.
“By teaching an AI how to create images,” Hu said in an artist’s statement, “I’ve learned a lot about my style—what makes it unique and where I, myself, might evolve.”
No Comment! Be the first one.