The bad parts are tales—told by students themselves—of getting tricked by AI, or, accounting for free will on the side of the student and a (purported) general lack of consciousness on the side of a large language model, tricking themselves into believing falsehoods. Not falsehoods in the sense of “The Earth Is Flat!!!1!” but smaller falsehood that sting more like “I understand this topic” or “I know this programming language.” I am reminded of Feynman’s famous quote:

The first principle is that you must not fool yourself — and you are the easiest person to fool.

It is clear that, to some extent, we are fooling ourselves. We are fooling ourselves by performing cognitive offloading at unprecedented scales. We are fooling ourselves by accepting shallow summaries instead of the nuanced whole. And we are following ourselves by believing that we have to do it because everyone does it.

— Higher Education and AI: Some Musings by Bastian Grossenbacher-Rieck