I saw a common theme in teachers’ posts about AI: “We’re on our own here! Why doesn’t the administration set clear guidelines?”
Viewed in the abstract, it seems to be a compelling question. But from “we need a policy” to “we need The Politics” is a lot harder than it sounds.
That’s partly because AI is a fast-moving target. In just a few years it has gone from a laboratory model to a seemingly obligatory future, even if it functions as a paper vending machine on site. It moves much faster than the policy development process can, so it is likely that any policy will be out of date by the time it is adopted. Yet it is everywhere; Just last weekend, my younger niece, who is in the middle of high school, confidently informed me that “everyone uses it.”
Some employers seem to want AI skills even though they don’t seem to know exactly what that means, and some companies are desperate enough to find students who haven’t outsourced their thinking because they’re hiring liberal arts majors and training them in technology. It’s a return to a much older model – and frankly, a welcome one.
A “Great Refusal” policy would be really attractive to lecturers who are (rightly!) tired of reading sloppiness passed off as real student work. But it would potentially put students at a competitive disadvantage when competing with others who have done compelling work at it. It also assumes that AI can only be used for bad purposes. Considering how quickly it’s evolving, I’m reluctant to suggest that it can only ever do harm, even though I see enormous harm being done right now. That would effectively rule out innovation. (I’ve heard of instructors having their students proofread Slop as an editing exercise; there’s something to that.) And it would force instructors to enforce rules that are so widely ignored that the legitimacy of enforcement is called into question. When culture and rules grow too far apart, culture wins.
The environmental objection to AI is real at this point, although I don’t know if that will always be the case. I could imagine that data centers would become much more efficient over time, then the environmental objection would be moot. If the real objection is being too sloppy, then resist the slop.
(Personally, I object to AI’s co-optation of the em dash. The em dash is a perfectly honorable writing tool. As Elle Cordova memorably put it, just because it’s stuck in ChatGPT’s teeth doesn’t make it bad.)
A large rejection would lead to serious problems in computer science education and, increasingly, in related health programs. Since it becomes part of the practice of the areas we prepare students for, omitting it would leave a gap in their preparation. It could even put ethics teaching in a difficult position by requiring students to make judgments about something they could not try themselves.
The “you can use it for suggestions and structure, but not for content” policy is probably far too nuanced and subtle to be enforceable.
A laissez-faire policy is, in effect, no policy at all.
The policy of “It’s the future, so let’s encourage students to use it” practically leads to a devotion to paper-omatics. An old joke compares artificial intelligence to natural stupidity; My concern is that the former leaves the latter undisturbed. The point of writing, for example, a five-page essay analyzing a marketing campaign, a social movement, or a novel is not the essay itself; It is the process of creating the paper. That’s where the learning takes place. When the process is reduced to a prompt, the learning stops there. It’s the difference between running a marathon and running a marathon. Yes, riding it is faster, but it won’t get you in shape. Allowing marathon runners to drive defeats the purpose of a marathon.
A policy that allows each academic department to set its own policies has the advantage of potentially differentiating computer programming courses from sociology courses. But it also creates incoherence on the ground and a potential race to the bottom as departments compete for enrollment. And for very small departments it’s not much different than having no policies at all.
All of these assumptions assume that the way AI looks will not change significantly for years to come. Given what we’ve seen in recent years, this is a poor choice.
Currently, many institutions are of the opinion that it is better to wait for the dust to settle before making blanket statements. For someone upset about the paperwork, I can understand why this might seem like a surrender. But filling the gaps of actual policy – when it is pervasive on the ground and the mechanisms are changing – is somewhere between ambitious and presumptuous. And as I say this, I join the holdouts and declare that you can pry my strokes from my cold, dead hands.
https://www.insidehighered.com/opinion/columns/confessions-community-college-dean/2026/07/24/fumbling-toward-ai-policy
