* Apple is able to charge a premium on iPhones relative to the best non-Apple alternatives because a lot of people are willing to pay it. (Do they have a monopoly on smartphones? Definitely not; there are loads of companies selling good phones. But they have a monopoly on their smartphones.)
* Fast food joints are forced into heated low-margin competition with each other because they're pretty interchangeable. If McDonald's starts charging too much, people will just shrug and buy basically the same food from Burger King.
* Classic monopolies are extreme cases. At one point if you wanted telephone service in the US you either paid whatever AT&T was charging or you didn't get telephone service.
When people hear Thiel's comments they tend to think of classic monopolies, but most of the examples he gives are of companies that have plenty of competitors -- and a durable advantage that they can use to maintain decent profits margins without customers jumping ship.
Sure you could use one of the alternatives, but then you won't get the content/people/apps/audience/... that are on those large platforms, which makes them basically useless.
Energy, agriculture, transportation - all built on the unintuitive building blocks of physics, and our approximate understanding of the universe. The clean abstraction layers and separation of concerns in IT is unparalleled, with powerful positive feedback begetting further innovations.
Applying advances to other fields also grew economies of scale rapidly, supporting innovation while barriers to entry were low. We're now approaching the end to low-hanging fruit and maturing into the consolidation phase, as further innovation requires greater capital and time commitment.
I spotted this (IMO badly mistaken) view of his when he first came out with it quite a while back:
Pay to drive the premium roads when you’re in a rush, or take the discount lanes with advertising printed along the roadway.
We could have quiet, porous asphalt with better water absorbtion. So many fantasies, I digress.