So there’s this thing I read called the “AI paradox” or something.

Once an AI gets good enough at a task, people stop calling it AI.

I was thinking about that too while reading it! Neat framing on their part. As someone who studies the field, I'm pretty torn on usage of "AI" myself - I think I like ML more, but as the capabilities of these models advance, it's hard not looking at their functions as some form of "intelligence".