Oct 23, 2022
Mixing AI Art Tools, Dilla Time & The Changing Same
Lately, I’ve been mixing output from more than one AI tool. In the image of Sidney Poitier (above) I started with text prompts in Midjourney. Then, I used the output as a style for another image in Deep Dream Generator. I composited the results using Photoshop and extended the background using DALL E 2. The result captures everything I had in mind and reminded me of hip-hop music production, specifically sampling snippets of different sounds to create a new work.
Ethan Zuckerman wrote a really cool article in response to recent criticism of AI-generated art AKA “Art is dead.” In the article, Zuckerman compares AI art to hip hop, specifically the innovation of J Dilla (‘Dilla Time’):
Dilla tapped his beats into his Akai, then nudged the sounds, microseconds at a time, into configurations not even he could play manually. The resulting off-kilter, drunk-sounding rhythms have been so influential that Charnas argues for a third rhythmic feel, beyond the “straight time” of classical music and drum machines and the “swing time” of jazz and blues: “Dilla time”, a rhythm only possible at the meeting of human and machine.
Dan Charnas’ book “Dilla Time” is a great read by itself but the comparison to AI art resonates with what I do that, according to Zuckerman is at the “intersection of AI and human skill.” I spent countless hours, days, and years experimenting with Deep Dream Generator (since 2017). By the time Midjourney and DALL E 2 came out I had DDG-generated images in exhibitions and even as an outdoor mural in Brooklyn.
Earlier this year I had AI art on display at the Smithsonian, Carnegie Hall, and Lincoln Center. My portrait of Greg Tate was part of a deep-funk second-line procession of drums and horns at a NYC SummerStage event. The image was scaled up, printed, and held up by led by bassist Melvin Gibbs. In the case of the second-line parade, the community was part of the exhibition. For me, this is just as important than having static work on view in galleries, which are often far removed from certain communities.
Amiri Baraka wrote that community persists in and as a conversation that continually refigures its past and future through what he called the “changing same.”
Within that American experience is the history and life of the African American Nation; a piece of the whole… whose only forward direction must be toward Self-Determination!
For me, being here has always been a condition of struggle and, hopefully, growth.
My AI art works out the relation between vernacular and technological determinism that posits that new technology directs and causes changes in society and culture. I think my purpose for making the art pushes back against some of the criticisms of AI art such as ‘Is it art?’ or ‘Who is the author?’ My work, more specifically, my willingness to share the processes I use to create the work addresses these questions.
My cover art for Ibi Zoboi’s upcoming YA novel “Nigeria Jones” (above) is a kind of ‘exquisite corpse’ in that the base image is a combination of three stock photos, to meet the character requirements from the author. I also used DDG to generate multiple images (from the base) and composited them in Photoshop. For some the use of AI to make art is far removed from traditional notions of art making. However, I see no difference between what I did for “Nigeria Jones” and paper collage or MPC sampling.
