Evaluate use of C2PA to detect AI-generated images

[nicolas-raoul]

Similar to how we detect pictures downloaded from Facebook and educate the user with a confirmation popup, ideally we should detect AI-generated pictures and educate the user with a confirmation popup (something like "This image was AI-generated, do you really want to upload it?").

Maybe https://proofmode.org/blog/simple-c2pa could be used for that? It implements https://en.wikipedia.org/wiki/Content_Authenticity_Initiative