This. If everyone in the tech industry adopted a "X works and many people use it, so we shouldn't try to make Y work" then you can say goodbye to progress. Specifically, some of us don't want the internet to serve as an interactive billboard, where our interactions with billboards are monitored. 99% of users probably don't care but being part of the next generation of software engineers I want to build things that influences people lives more positively than the previous generation of tech. I honestly question the moral integrity of my fellow technologists. There's a massive gap some commenters are missing between looking at the outcome of a model and the technical or conceptual changes it could undergo over time. Just because something has been working for 5/10 years doesn't mean it's the only proven way nor do we need to enforce it as the "proven ©" status quo for everyone else because we're too afraid of technical changes.

I don't think you can place the burden of evidence on one side alone. Enough people have experience with YouTube as users, creators, or both that it's not wild to think that many people have reasonable intuitions about why YouTube dominates the market.

And if we are going to demand evidence of critics, I think we should also demand evidence of promoters of things like this, who often seem guided by a Big Idea rather than any evidence of user needs or proven ability to solve those user needs.