You can represent many things with numbers and build an algorithm that does stuff. ML techniques are formulas where some specifics constants are not known yet, so you go through a training phase to find them.

While combinations of words are infinite, only some makes sense. So there’s a lot of reccurent patterns there. When you take a huge datasets like most of the internet and digital documents. I would be more surprised if the trained model where incapable of producing correct texts as both the it’s overfitted to the grammar and the lexicon. And I believe it’s overfitted to general conversation patterns.