I swear this is the real reason AI companies are concerned about watermarking - not for the public good, but so they can keep their model training inputs clean.

Copycats might also be an issue. I remember that when Douglas Wreden published his book, he had to deal with numerous AI-generated copycats popping up one after the other, using his name and face, trying to trick people into buying their book, because his had become popular.

I get the impression that most kindle books are sold (well rented) by subscription, i.e. all you can read for a fixed monthly fee, so a subscriber downloading an AI book or other crap doesn't pay any increment to do so. IDK how the revenues are divided up.

Before AI they offered a lot of books which were just printed wikipedia articles.

However AI books massively increases the problem, because they can be created by the thousands, millions even, in seconds by any grifter out there, and there are a lot of grifters.

Any sort of argument of "it has always been bad" (and it has), are not very constructive, because the current situation enhances the shittification of the internet on a scale that was impossible before. It is an unprecedented situation.

AI will never become JARVIS, KITT, HAL9000, or whatever super AI you seen in media.