generative art vs ai art: what's the difference?
they both make images with machines, so people use the words interchangeably. they're not the same thing. here's the difference, plainly.
the short version
generative art is made by a system following rules the artist wrote: an algorithm, some controlled randomness, a set of constraints. the artist designs the process, and the process makes the image. ai art is made by a trained neural network, usually a diffusion model, that generates an image from a text prompt after learning from millions of existing pictures.
side by side
| generative art | ai art | |
|---|---|---|
| how it's made | hand-written rules / algorithm | trained neural network |
| your input | parameters, seeds, selection | a text prompt |
| repeatable? | usually deterministic, same input, same output | often varies run to run |
| training data | none | millions of existing images |
| examples | flow fields, cellular automata, gaogao | midjourney, dall·e, stable diffusion |
where gaogao sits
gaogao is generative art. every organism is grown deterministically from a genome, genes with dominance, so the same genome always grows the same artwork, on any device, forever. there's no model and no prompt. what makes it unusual even for generative art is that the pieces are genetic: you can keep the ones you like and breed two of them into offspring that inherit a mix of their traits.
so if ai art is "describe it and the machine paints it," generative art is "design the machine and watch what it grows." gaogao is firmly the second, with a little biology in it.
grow generative art, free →