Stable Diffusion

For example, with Holly Herndon and Mat Dryhurst's project for the Whitney, it uses, I think, Stable Diffusion under the hood, but it's got this LoRA or fine-tuning that they did on it so that when you type fluffy bunny into the interface, it doesn't give you a fluffy bunny. It gives you Holly Herndon as a fluffy bunny.

Jon Ippolito, Preserving Digital Art: Regina Harsanyi, Jon Ippolito

I was interested mostly at the time in variational autoencoders, and that's a generative network that still exists now, kind of inside of models like Stable Diffusion.

Tom White, Tom White & Gene Kogan on the Birth of GAN Art