
When the instructors became aware of the extent of this issue, they reached out to approximately 100 students who appeared to be engaging in dishonest practices. “We contacted them with a cautionary note, asking them, ‘Please clarify your recent actions,’” mentioned Fagen-Ulmschneider in an Instagram video regarding the matter.
Responses from the students began to trickle in, initially few, then overwhelming. The professors felt encouraged by this display of accountability and remorse… until they discovered that 80 percent of the responses were remarkably similar and seemed to have been produced by AI.
Thus, on October 17, during their lecture, Flanagan and Fagen-Ulmschneider addressed their class, showcasing a collage of the responses, all featuring the identical phrase “sincerely apologize.” No punitive measures were imposed on the students, and the situation was treated with a light touch—but the warning was serious. Cease this behavior. Flanagan expressed a desire for it to serve as a “life lesson” for the students.
Time for a life lesson!
Credit:
On a University of Illinois subreddit, students recounted their own stories from the same course and of AI utilization on campus. One student asserted that they were a teaching assistant for the Data Science Discovery course and noted that, apart from not attending, many students relied on AI for the (fairly simple) assignments. AI tools frequently “utilize functions that weren’t covered in class,” making it evident.
Another TA stated that “it’s astounding how ubiquitous AI-generated content is in 75% of the submitted assignments,” while another student expressed frustration as a course assistant where “students were required to submit a 75-word paragraph each week, all of which was AI-produced.”
One doesn’t need to look far in these types of discussions to uncover numerous students who feel wronged because they were accused of employing AI—but hadn’t actually done so. Given the limitations of most AI detection tools, this scenario is quite believable; and when AI detectors aren’t utilized, allegations often hinge on intuition.