They do build a representation of words and sequences of words and use that representation to predict what should come next.
A simplistic representation is this embedding diagram that shows how in certain vector spaces you can relate man/woman/king/queen/royal together:
The thing is, these are static representations and are only bound to the information provided to the model. Meaning there is nothing enforcing real world representations and only statistically consistent representations will be learned.
Sorry I've got to down vote you, but its in your own article:
Hard to link to the specific line, but this is only one of a few instances where they state a passport is enough.
Edit: If a passport wasn't enough, the percentage impacted by the bill would be a lot higher.