> Is widespread, mandatory watermarking the only answer, or is it a technically flawed/doomed approach?
I don't see how you could seriously trust a technical solution like that. All technical solutions can be pretty much broken. In my opinion, you also need to build up a web of trust of people who are definitely against AI and who make it prominent that they refuse to use any generative AI content. For example, on my YouTube channel I make that abundantly clear, and on the website I work for we also have an anti-AI statement.
A better "technical" solution would be to build a platform that attracts like-minded people. Yes, once it gets large it might be flooded with AI content/bots, but the trick is not to let it get large, nor go for a large profit at first along the lines of typical technology solutions. The internet is flooded with AI content not because AI suddenly and rapidly came to exist, but because there was a niche for it.
Personally, I hate AI, with a passion. Find people who create and write and who hate AI and it's a start.
