The Honor-Roll Offloaders
The students leaning most heavily on GenAI aren't the ones failing. They're the ones with the highest GPAs—and that should worry us more.
Think about the student you expect to be the last one to cheat. Not the one looking to save themselves with a C, but the student with the 4.0 GPA, getting recommendation letters with ease. Now, picture her history with ChatGPT.
In the 2026 preprint “Patterns of Student Cognitive Offloading to AI in Higher Education,” researcher analyzed 3,046 ChatGPT messages from 46 college students. It found that the students that used the tool the most were not the low achievers, but the high achievers. It was the honor students, in this case, the most. (Although this study is a preprint, not peer-reviewed, and the sample is small, it does have value.)
This changes that narrative that we have. Many of us have the same fears about the new campus genAI that student cheating comes from a place of panic, and that the student is trying to achieve something they know they can’t achieve on their own. That student does exist. But, the data are showing that for the students that are doing the best in the class, this new dependency on genAI is not a panic driven move, but a way of thinking that has been refined to be the most efficient.
We built the system; it’s only fair that students try to navigate it. Students have been told that the goal of education is to get the grade, the prize is the transcript, and the end result of an assignment is the artifact you hand in. When students hear that, they have been trained to follow the logic, and a rational student will do the same thing. If a student knows that their goal is to produce an artifact, of course they will try to do that in the most efficient way possible. Right now, the most efficient way to do that is to use AI, which the student knows very well.
Students using AI hurts the students who it is really meant to help. Students who use GenAI do not need to use it the same way, because students who use GenAI do so because they are just being overloaded from so many different places. It’s a bargain that gives students back hours that GenAI stole and asks very little in return. Only those students who were overloaded the most from the education system use it, and very few have told them about the invisible costs of GenAI because it is still very new and being tested.
And it looks like these expenses are easy to pay later. The thinking that seems to be skipped shows no apparent gaps. The essay and the grade look fine. But the gaps appear years later. They are unclear and untraceable. They will come back to haunt you in adulthood when you are able to write documents with the best of them, yet you cannot differentiate between a good sound argument and a nice looking one, when the moment calls for it. You’ll get the bill without a name.
Now, when you view this problem from the registrar’s lens, it becomes something that is no longer an anecdote, but a transcript. This spring, a team from the University of California, Berkeley analyzed 500,000 sets of grades over the span of 8 years. In writing and coding courses, they noticed an increase in A grades by approximately 30%. This was almost completely due to an increase in A grades for take-home assignments and not proctored exams. As one of the study authors put it, you now have a C student who is now an A student. The grade went up, but the thinking didn’t change. EDUCAUSE Review called this phenomenon in 2025 “better results, worse thinking.”
So here’s my claim. The honor-roll offloader is not a discipline problem; she is a measurement problem. We spent a century building an education system that grades outputs, because outputs are easy to grade and were reasonably hard to game. GenAI has revealed that a graded output is no longer evidence of anything happening inside a head, and now the transcripts say so at scale. Our best students were the first to notice. Of course they were; noticing is what made them our best students.
The institutions have noticed too, and their alarm is the tell. Princeton is ending 133 years of unproctored exams; Harvard’s faculty voted to cap A’s at a fifth of each class. Both are confessions that the transcript has stopped certifying what it claims to. But capping grades treats the symptom. The real question is the one the Berkeley author asks: do grades still tell us what students can actually do?
What we should be developing is reasoned judgment: the capacity to sit with an argument’s claim and reasoning and decide whether it holds. You cannot offload that to a machine, because the machine was trained on the internet, where reasoned judgment was never required to post. Outsource your thinking to ChatGPT, and you become more reliant; learn to interrogate what it hands you, and you become harder to fool. Same tool, opposite results.
The fix isn’t detection; that’s a tax on trust the cleverest students will always outrun. It’s assignments where overreliance carries a penalty, and the thinking is the deliverable: done in front of you, explained out loud, and graded explicitly, as the thing itself. If we make the thinking the deliverable, the shortcut stops being short.
Until we do that, the most capable people in the room will keep making the most rational possible choice. We won’t like where it leads them, and for now, the transcript won’t even tell us it happened.
A question for the comments: If your best students are the heaviest AI users—and the grade can no longer reveal who learned what—is that a cheating problem or a curriculum problem, and what would you actually change?
Read deeply,
Dr. Genevive