I use a variation of Conti and Caroland (2011)’s “memorize the first 100 digits of pi” assignment as an in-class learning activity to teach my infosec students how to cheat, as part of teaching threat modeling. This semester at the beginning was all-remote for me, though, so I modified it for the zoom teaching session. Instead of having all students write down the first 100 digits of pi and then submit their papers to me, I instead told students the following:
Your assignment is to memorize the first 100 digits of pi having been intentionally given so little time to do so that your only chance of completing the assignment is to cheat. I authorize you to cheat on this assignment and only this assignment, but if I catch you cheating, you lose.
It will go something like this (subject to change): at the start of the next class, I will randomly sort the class roster, and then I will go down the list and call on people one at a time to recite to me the first 100 digits of pi. I may ask you to share your screen, and also to pan your camera about the room (be prepared!), simulating what proctor-monitoring software would do. The grading is pass/fail. If we run out of time before you are called on, you pass by default.
Again, I expect you to cheat. How you choose to cheat is entirely up to you. Collaborative cheating is also encouraged, but everyone involved will lose the game if caught. At the completion, I will give you opportunities to share-and-tell your cheating techniques with the class.
The objective of the exercise is to learn how an adversary thinks and operates by deliberately loosening traditional rules and tapping personal creativity. It is an exercise in threat modeling.
I am only permitting you to cheat on this single quiz. You may not cheat on any other assignments for this class.