I feel bad to say this, but this past academic year has probably been the one where student writing has bored me the most. Bachelor and master theses, take-home assignments and, the worst, motivation letters. Yes, the content was more “polished” than other years, but it all felt dreadfully the same and soulless. Grading is rarely an academic’s favorite task but this year, while grading, time seemed to drag on even more slowly than before.
Sharing this experience with my colleagues, I learned that I wasn’t the only one with these feelings. The culprit, we think, is generative artificial intelligence, or gen-AI. Students rely increasingly on gen-AI to “improve” their writing and, in some cases, to generate entire blocks of text (this year, I had to report several cases of students who used gen-AI to write most of their assignments). University guidelines on the use of gen-AI are continuously updated, and students have told me that they find it hard to know what is the “correct” use of gen-AI and what crosses a line.
Regardless, what bothers me is the idea that we are sacrificing the “soul” of writing (and yes, even scientific writing can have soul) for qualities such as grammatical perfection. Especially when it comes to pieces like motivation letters, I’m not looking for perfectly written letters and generic statements about enthusiasm for my research topic. I want to know about the person writing it and who I will be investing my time and effort with over the coming months.
Research currently shows that most people are not very good at guessing whether a text has been made by gen-AI. Among the general population, rates are about 50-50, no better than a coin toss. Interestingly, accuracy rates among university educators are often higher (up to 70 percent) and I swear that we have a sixth sense for it. Beyond distinguishing whether a single piece is AI-generated or not, our feeling that across students’ work, everything feels “same-y,” is a signal that something is seriously amiss.
Of course, I am writing from the perspective of the university teacher, and the question is whether it should matter what our personal experience is of reading students’ written work. For the sanity of me and my colleagues, I think that enjoyment and interest in our work is extremely important. Further, we care about educating students who cannot only produce well-written, meaningful work, but who also care about and deeply understand what they have written, both within their education program and beyond.
Jessica Alleva, assistant professor at the faculty of Psychology and Neuroscience