‘We Cannot Choose to Become Idiots’: A Brown Professor’s Proof of Mass AI Cheating
An economics professor at Brown University suspects most of his class cheated with AI, and he has the numbers to make the case. Roberto Serrano watched his take-home midterm average hit 96 out of 100. When he switched the final to an in-person test, the average fell to 48. He has taken the story public, telling El País and Inside Higher Ed that he will not let it go.
The Take-Home Format and Its Gap
The take-home format came about due to a tragedy—a gunman killing two students on campus last December. Many expressed anxiety about sitting exams in a room full of people, leading Serrano to offer take-home midterm and final papers as an alternative. Irony stings: the one time in decades he relaxed the rules, much of the class cheated.
The Numbers That Gave It Away
Serrano’s course, ECON 1170, usually draws a small, strong group. This term, 86 students signed up—likely drawn by the new take-home format. The midterm results were extraordinary: an average of 96, with 40 scoring perfect 100s. Historically, the course averages between 65 and 80.
The answers themselves felt off, with many correct ones sporting a "very convoluted style." When Serrano and his graders fed the questions to ChatGPT, they received similar responses.
Setting a Trap for Cheating
Serrano decided to compare the midterm and final distributions. If they matched, he would keep the midterm; if not, he would void it and reweight the final. The response was clear: 18 students dropped the course, and 9 more skipped the final. Of those 27, 22 had scored a perfect 100 on the midterm.
The final average dropped from 96 to 48. Serrano estimates at least 50 students cheated on the midterm, calling the evidence overwhelming.
A Wider Reckoning
Brown is not alone; a recent Princeton survey found that 29.9% of students admitted to cheating on at least one exam or assignment, often using AI. Schools have been chasing detection tools and rethinking how they test for two years now.
Students feel the strain too: Brown’s own provost-led report revealed that most undergraduates use generative AI weekly or daily, yet large majorities worry about its effect on their learning and cognitive capacity.
Serrano frames the issue starkly: "We cannot afford to have a society in which a significant fraction of our best young minds think that cheating is acceptable."