This is great! I hope somebody adopts and publishes such a package.
There are some pieces of widely used software (numpy, git, regex...) that are very powerful but can be highly complex, to a point where reading them is much harder than writing them with trial and error. I hope we see more projects like this, that add layers on top of those, to make them readable without making them less performant.
(Such projects likely exist already, of course, so they just need to be polished and spread)
1. They don't need an enormous number of artists; the research paper showed significant results with even 50 poisoned image samples in the dataset, which is enough to be contained in even a single artist's online gallery.
2. They don't need to keep it a secret; the goal is to remove these images from the training data, in a way that would be much more efficient than simply adding a "please don't include my art in your ai scraper" message next to your pictures.
Another pitfall: this is easily circumvented by the end user generating a long text and then randomly adding/removing a few words here and there. This could be solved by changing the simple check of "every Tth token belongs to S" to something like "the average distance between subsequent S tokens is very close to T".
