Yeah, I don't think people understand quite how astronomical an undertaking it is to replace this shit. People like to quote things like AWS, but AWS is a) expensive and b) general purpose. As such, it might be able to solve the problem, but not nearly as efficiently. It would cost you proportionally WAY MORE than Google is paying to keep YT alive, so that gives you an extra giant hurdle on top of the other complexity.

Web hosting with low latency is hard. Huge data storage is hard. Transcodinf is hard. Constant uptime is hard. Search is hard. Recommendations are hard. Making it profitable is hard. Starting an ad service that isn't googles is hard. Convincing content creators to move there is hard. Convincing consumers to look there is hard. Sure, any of these problems have remotely comparable analogs. But you have to solve all of them simultaneously to get anywhere near competing with YouTube. And since Google owns the whole "stack", it's much cheaper for them then it'll be for you.

Kick probably makes a decent comparison here. But they're A) solving a subset of the problem B) fighting against a company that has extremely clear problems (arguably much worse than YouTube) C) is in a tech savvy-er demographic D) is funded by mega-casinos with tons of money and a vested interest in the product E) fighting in a market with less inertia so viewers and creators can move easier F) fighting twitch instead if YT which is smaller and younger.

And they're still not really all that much competition.