To start with the last first, I would argue your general (and not uncommon) thesis is off in one key respect: it's not primarily (though in some percentage it may be) a matter of people building
where they shouldn't have per se, but that they built
what they shouldn't have
where they did. Ie, it's 100% feasible as a matter of architecture/engineering/construction to build a structure that will shrug off a Cat 5 hurricane including storm surge. And while it adds a real premium, it's also not at some impossible cost either. People have done it, and it works. Same with most other natural disasters. It's "only" a matter of cost and standards. It's worth noting too in many cases societies have indeed done exactly that, like with earthquakes, or in areas of high risk tornadoes. Building standards have been set to match the risk. There is plenty of low hanging fruit that can severely diminish the impact of a lot of the disasters causing massive damage if it's just standard upfront.
Also, there's the complete polar opposite approach: build something "disposable". In the "old days" (including with my extended family) there was a style of "summer camp" for example that was ultra simple. Some small single floor deal, uninsulated, maybe some power but often not even that, composting/pit toilet, some simple wood furniture, that's it. People bring their own everything, be there for a few weeks/months a year, and then go home. Such a structure can't survive much of anything but that doesn't matter because it's so cheap, if it burns/blows down/washes away once every 5 years or whatever so be it. It's a problem though when people convert what should be cheap into some full fledged thing, but then don't take environment into account.
I think this distinction is super important, because a lot of these places are beautiful and desirable much of the time, and a blanket "no you shouldn't build there ever" isn't likely to be heeded and does not get to the root actual problem, which is that the true costs of doing so aren't being priced in. The reasons for that distortion are myriad, but that's the actual issue. I think it's much more productive and convincing to the public to say "it's fine to build where you like, but it's not fine to hit other people up for money to cover it or cause unreasonable costs to safety services/environmental damage (homes burning or floating away means massive pollution), you just need be responsible in how you build."
FWIW to specifics:
>With much of California, Texas and Florida it seems pretty clear people built where they shouldn’t have.
In some cases sure but in others I guess it'd be reasonable to say that things built long enough before anthropogenic global warming really kicked off can't be reasonably blamed for that, particularly if they correct gauged the risk for themselves (ie, someone built something 50 years ago as a life thing and it did indeed last the remaining 40 years of their life, well you can't really say they got it wrong and built it wrong or it's still their problem). What is bad though is new stuff getting built or worst of all things getting REbuilt after destruction but not to updated standards each time.