When you start getting into the specifics, it becomes way more complicated. How exactly should these AI companies notify people that their content is being used for their model? First of all, they're not actually the ones harvesting the data. That scrapers tend to be independent... so these artists are going after the wrong people, unless you expect the AI company to parse through all the data they use to find the rightful owners of everything and ask for their consent, which isn't really viable, let alone practical. Let's suppose the artists do go after the scrapers, how exactly do they notify people that their content is being used? The content is collected by an algorithm, how are they supposed to reliably identify the rightful owners of content and ask for their consent? Do they just send automatic messages to any email or phone number they find?
How about this, what if an artist is posting their art on a platform, like say for example Reddit, and that platform agrees to allow the data to scraped and used for AI data training? Does the platform company own the data on the platform or the individual artist? If it is the latter, what's stopping platforms from modifying their TOS to just claim ownership of anything posted on their platforms? Again, what is the ultimate goal here?
The point is that while I agree that AI has to be regulated, the criticisms and proposed regulations have to specific and pragmatic for them to mean anything. This general hatred of AI and whining by artists and other groups is just noise. It's just people trying to fight against technology, and as history has shown us before, they will inevitably lose. New technologies have always threatened and displaced well established workers, careers, and industries. For example, lamp lighting used to an actual job, but as the technology improved and light bulbs became a thing, lamplighters became a thing of the past. They tried very hard to resist the change and managed to do so for awhile, but it was a losing battle and they eventually faded away. Economics and technology always win.
That's kind of the key here, these generative AI's are the light bulbs of our era. They've already replaced a bunch of jobs and radically changing entire industries. There's no ultimate goal with them and there's no fighting them. Pandora's box is open and it's not going to close. This new technology is still at it's infancy now, but it's going to rapidly expand, evolve, and adapt to a bunch of different situations. Whle regulations can help guide this freight train of a technology in the right direction, they can't stop something with no brakes. As it gets adopted by more and more people and used in more and more spaces, it's going to alter how we do things kind of like how smartphones or social media did. We have no choice but to evolve with them or else we'll become the new lamplighters.