It could be a bit dangerous to rely only on those when building your startup, but having a good combination of these early adopters and pre-sale conversations with late adopters is super important.
Zero audience is not a problem btw since a) there’re are usually relevant communities, b) there’s trivially automating cold outreach that is a single best way to test your PMF hypothesis.
Edit: typos
In this case, I think there's a strong argument to be made that while yes, there are people who will be willing to buy into an unknown startup's unreleased product, by and large, those people are fools (in the sense of "a fool and his money are soon parted", at the very least). There's certainly a segment of our population (which is, completely unsurprisingly and reasonably, overrepresented on HN) that's very excited about startup culture, and wants to believe that any given startup is going to Change The World™...but far, far too many of them fall into one of three categories:
1) Come up with a "new" way of moving money around so that rich people get richer (eg, the entire cryptocurrency space)
2) Make something that looks just flashy enough for them to get bought out by a bigger company
3) Fail outright
I would do that if I already trust the organization, which implies it has already launched useful products in the past. For an unknown startup, the idea would have to be truly Earth-shaking for me to sign up.
In most cases, I'll just wait until I hear that the app is out, and then try it and consider purchasing.
To me, "sell before you build" is a no-go for upstarts with zero audience, whereas "market before you build", or maybe "market as you are building", creates the awareness you'll need to warm people up when you are ready to sell.