Wow! I just wrote the article "Stop trivialising AI: it’s not your intern, colleague or compiler" today and shared on HN - https://news.ycombinator.com/item?id=47274213 - but I talk about different dangers though. Enjoyed reading your viewpoint.
Exactly right. The author does miss something that journalists seem to like, "AI" apologizing. LLMs can print words that if a human said them, would constitute an apology, but an LLM is a program and a context, so that isn't an apology.