Why would one need an AI to generate a convincing time-lapse video of one's work being drawn and painted line by line? Did you forget to record one, artist?
Paints-Undo is a project aimed at providing base models of human drawing behaviors with a hope that future AI models can better align with the real needs of human artists.
The name "Paints-Undo" is inspired by the similarity that, the model's outputs look like pressing the "undo" button (usually Ctrl+Z) many times in digital painting software.
Paints-Undo presents a family of models that take an image as input and then output the drawing sequence of that image. The model displays all kinds of human behaviors, including but not limited to sketching, inking, coloring, shading, transforming, left-right flipping, color curve tuning, changing the visibility of layers, and even changing the overall idea during the drawing process.
An answer is "because you want to conceal the fact that you used AI to create the finished work." And why would one need to to that, exactly? Because requesting timelapses is a key proofing method right now to detect AI use.
As it stands, from the example, I don't think it'll work too well for the presumed purpose of fraud, as it generates a much larger set of images all liable to contain AI "tells." The venn diagram of intent, execution and opportunity are not yet a perfect circle.
I'm inclined to a liberal attitude toward AI tool use by individual artists—it's the companies and their economic and political power to be concerned about, and making whole those they exploited to train the tools in the first place. But, obviously, artists shouldn't be lying about it. And even if you're honest, you have to accept that some find it in bad taste.
A second-order problem of AI tool use is that it's still extremely difficult to replicate details, forms, models between generations. Asking for changes in pose and scene quickly exposes the fakers.
AI boosters often post examples of "perfect" persistence of character and scene design, only for these examples to be full of countless differences obvious to others. This leads to comical outcomes on social media. But it hints at something ominous: as a product, generative AI doens't have much incentive to improve beyond a certain point. It will function similarly to spam emails from foreign princes and bankers: the flaws and obvious absurdities repel savvy operators while attracting self-selecting victims.
Previously:
• Dune subreddit bans AI art
• Illustrator discovers her art was used to train an AI art generator
• Artists upset after Wacom uses AI art to market artist gear
