Can we see how it plays out a bit first before we start tearing up the carpet?
I think some people are too fast to read the worst possible outcomes as soon as they see "AI", and the change you're recommending is massive, so let's deal with some of what's actually, materially happening here.
The battle over AI will play out and ultimately AI content will be allowed. That's honestly the only eventual outcome. I can see the problems with that, but I think a lot of people are ignoring how hard it'll be to stop. The genie is out, the horses are gone. Our mission is to help Ubuntu users, not chase out AI posts.
The way SE Inc treats its community is a factor but it's not completely unlike the way Canonical treats its members and users. Sometimes companies rule their communities by decree and in their interest. I hold this in balance with what we get back.
This system isn't just what you see, there's a network of spam control infrastructure and moderation tools that benefit hugely from the scale of the whole.
Prosus didn't need to buy SE to train "their AI" on SE content. It's been available to anyone for a long time. And given their portfolio, I'd fully expect them to offer more products based on SE content, because why wouldn't you? Others are.
Lessening the demand on volunteers isn't automatically a bad thing. Even if that gets down to the point where we're just review. Our mission is to help Ubuntu users.
I think there's plenty to be malcontented with, but on balance not enough to build and host an alternative elsewhere. Maybe I don't appreciate something as much as others do though, but I do have quite a good idea about the costs and resources needed to run something like Ask Ubuntu.
I realise you're only asking us to plan, and it's always wise to have an idea in the back of your head, but I think we are still several steps, several of these "what if they make it worse" away from that balance tilting to the point where we'd be better off elsewhere. And we're still not sure that these things will make it worse.