This just isn't true in my experience. 2026 LLMs, especially if prompted well, are much better at writing than most humans. I think journalists really underestimate how bad most people are at writing.

Here, Kriss correctly identifies this ChatGPT prose as being meaningless garbage. But spend any time on Tumblr and Reddit and you'll know that people write like this *all the time*.

Kriss gestures at this when he notes how much drivel even "real" writers put out, but he doesn't acknowledge how much this undermines his argument. (also, why does he get the title of the show wrong? it's "The Miniature Wife". is this one of his "ooh i blend fact and fiction!!!" affectations?)