Comment by zarzavat

The assumption here is that LLMs will never pass the Turing test for copywriting, i.e. AI writing will always be distinguishable from human writing. Given that models that produce intelligible writing didn't exist a few years ago, that's a very bold assumption.