General Discussion
In reply to the discussion: Everyone Is Cheating Their Way Through College (James D. Walsh, NYMag. Horrifying read on ChatGPT destroying education) [View all]HoneyAndLocusts
(1 post)Last edited Thu May 8, 2025, 09:09 AM - Edit history (1)
Higher ed instructor here.
No question, an instructor who tells students to "write" essays by tweaking AI-generated output has abdicated a core duty of their role. I tell my students that everything they write must come directly from their own brain, and this is the hill that I will die on. However, I understand why some professors choose surrender. In a few short years, students have become alarmingly reliant on ChatGPT (or as they call it, "Chat''). No matter how sternly you threaten or how sweetly you cajole, no matter how cleverly you design the assignment, if you let students do work out of class, a good percentage of them will turn in AI-generated garbage. I'm at a relatively prestigious institution, and if I ask students to do out-of-class writing (which I've discontinued almost entirely), at least a third of the work I receive is either unambiguously AI-generated or extremely suspect. And I'm sure some of the AI-generated material that I encounter flies under my radar entirely. If they copy-paste unedited ChatGPT output, it's glaringly obvious, but most of them are smart enough to hide their tracks. Horrifyingly, some of them probably even believe that they actually are doing the work when they fill in the blanks of an AI-generated outline.
At a community college, where many students arrive academically underprepared and juggle education with work, I wouldn't be surprised if a large majority of students used AI. Practically speaking, you can't report the majority of your students each semester for academic dishonesty: you will swiftly find yourself out of a job. So unless you are prepared to eliminate all out-of-class assignments, when you prohibit AI, you're making a rule that you have no way of enforcing, and then you have to grit your teeth all semester long as students flout that rule.
Another factor to consider: to my endless consternation, some institutions either encourage or outright require instructors to permit AI use. My own institution grants instructors a reasonable degree of latitude to set our own course policies, but there's definitely some worrisome pro-AI rhetoric coming from the administration; reading between the lines, they think it's the inevitable future and we have no choice but to get on board. My personal view is that climbing on board the AI train is like getting in the van with the kidnapper: if you don't fight like hell right from the start, you're losing your best shot at getting out alive. But alas, not all of my colleagues--especially the ones who have installed themselves in positions of institutional power--have the appetite for this fight.