I like the analogy to car theft. It communicates how shocking plagiarism should be within a healthy academic community. And stealing cars will normally attract

I like the analogy to car theft. It communicates how shocking plagiarism should be within a healthy academic community. And stealing cars will normally attract serious attention from the police. Of course, if you impose draconian punishments for a first offense, then you also increase the burden of proof. It's not enough to say that "student X cheated with 60% probability," any more than you could say, "It's reasonably likely that the defendant stole the car." You need to have clear-cut evidence. One of my friends was a lecturer in the computer science department (at the honors 200 level), and he made it clear that he took cheating as a personal affront. Every year, he'd catch 2 or 3 students cheating, and he'd devote considerable time to educating the student judiciary about CS and getting the cheaters suspended. Despite this, he was an enormously popular lecturer. But he was operating within a culture where plagiarism really was a serious offense against community norms. In a (very) perverse way, it is a matter of fairness to the students for me to be as vigilant as everyone else. Ultimately, this fight can't be won by a single professor. It's a matter for your colleagues, your administration and even your student body: Does NYU want to be a university where cheating is shockingly unacceptable, or one where cheating is on par with a parking ticket?