Yeah, I am sure enough about this not happening that I am willing to place bets. There is an enormous amount of intangibles Coursera can't give you (I agree it can be useful for a certain type of person for certain types of aims).

Agree that being inside academia is probably a lot bigger deal than people outside it really appreciate. We're about to see the first generation that grew up with a really ubiquitous internet come to grad school age though. Currently in addition to the assumption that generally clever people will want to go to university, we've treated it as obvious that the Nobel prize winning clever people will have an academic background. Which has been pretty much mandatory, since that used to be the only way you got to talk with other academicians and to access academic publications.

What I'm interested in now is whether in the next couple decades we're going to see a Grigori Perelman or Shinichi Mochizuki style extreme outlier produce some result that ends up widely acknowledged to be an equally big deal as what Perelman did, without ever having seen the inside of an university. You can read pretty much any textbook or article you want over an internet connection now, and it's probably not impossible to get professional mathematicians talking with you even when they have no idea who you are if it's evident from the start that you have some idea what their research is about. And an extreme outlier might be clever enough to figure things on their own, obsessive enough to keep working on them on their own for years, and somewhat eccentric so that they take a dim view on academia and decline to play along out of principle.

It'd basically be a fluke statistically, but it would put a brand new spin on the narrative about academia. Academia wouldn't be the obvious one source of higher learning anymore, it'd be the place where you go when you're pretty smart but not quite good and original enough to go it alone.