Every few months a new list of prompting tricks goes around. Most of them stop working with the next model. We spend very little class time on them.

What lasts

What lasts is judgment: knowing what a good answer looks like before you ask, noticing when a result is too neat, and checking the one claim that everything else depends on.

We train this with a simple exercise. Students get three AI answers to the same question, one of them subtly wrong, and have five minutes to find it and prove it.

Why it matters

The people who will do well with AI are not the ones who type the cleverest prompts. They are the ones who can tell the difference between an answer that sounds right and one that is right.