#15 (251030) Daily Experiment
a fusion of old (2021-2022) AI text-to-image tech (when it was still flawed and damaged and beautiful) with new (Oct 2025) Ai video tech.
original image crafted with Disco Diffusion 5.7. video generated with Grok Imagine 1.3.2.
when i make a work of analog physical art, i never start with a subject in mind. rather, i surrender to unconscious flow, and am often surprise by the finished piece. which is one of the things i enjoy most about creating art.
Disco Diffusion is the only AI text-to-image generator i’ve found thus far that fits my own work flow so closely. it’s an older model (circa 2022-23) that’s slightly unpredictable and uncontrollable, making it particularly enjoyable to use. more modern models tend to create images a little too predictable and refined for my personal taste.
and i am absolutely fascinated with the recent improvements in AI videos, in particular, with Grok Imagine. So much progress in such a short amount of time.
for a long time i avoided making videos, because i didn’t think i would like it. i’ve always been drawn to still images. but come to find out, creating them is fun as hell. who’d of thunk it ? …
e9Art (aka eric) has been a typesetter, tire changer, the coffee dude, that pizza guy, a homeless couch-surfer, a gag-writer, children’s picture book author, award-winning and internationally published haiku poet, fine-art photographer, and since 2016 a full-time artist (self-taught).
his first NFT collection was created in May 2019 (OpenSea), he minted his first work of AI text-to-image art in March 2022 (Foundation).
links to interviews, bios, social media, analog paintings and NFT collections can be found at https://linktr.ee/e9Art.