For the first time since he began teaching welfare economics and social choice theory nearly two decades ago, Brown University economics professor Roberto Serrano gave his students a take-home midterm this spring. A number of students had expressed fear of being in a classroom after a gunman killed two students and injured nine in a December mass shooting at Brown, and so “it was appropriate,” he said, to allow students to take their exams at home.
But at the end of the semester, Serrano regretted the decision. Dozens of students in the class likely used artificial intelligence to cheat and get perfect or near-perfect scores on their midterm, he said. Serrano, in turn, took the final exam in person, which resulted in more than a dozen students dropping out of the course and even more failing. Administrators’ response to the widespread fraud incident was “humble,” he said, and the incident raised questions about how universities can and should respond to AI-powered fraud at scale.
His social economics course typically has up to 30 students, but this spring he taught 86 — an increase he attributes to the promised take-home exams. At the mid-term exam, the average score was 96 percent.
“Historically, the average midterm grade for this course has ranged from 65 to 80 [percent]“And this exam was harder than the exams I’ve written in the past because … the take-home exam is an opportunity to make the course a little more challenging because you’re giving students unlimited time,” Serrano said.
He knew something “suspicious” was going on, so he and his evaluators conducted the test via ChatGPT. The AI gave answers that mirrored what his students had written and that were “kind of right, but very off and with a very convoluted style,” Serrano said. For example, one question asked students to prove a mathematical statement that could most obviously be proven with a “direct argument,” Serrano explained. ChatGPT – and many of its students – used a “contradiction argument” that provided the correct answer but was “very contrived” and that Serrano could tell was not written by a human.
In a message to students after the midterm he shared In Higher EdHe told them that he suspected many of them of using AI to cheat and, with his dean’s blessing, changed the final exam to an in-person exam.
“I won’t explain [the midterm] invalid for now. “I will give the class a chance to prove me wrong,” he wrote. “That is, if the final exam distribution is approximately the same as the midterm distribution, I will count the midterms.” Otherwise, and that is of course what I expect, I will declare the intermediate exam invalid and reconsider the final accordingly.”
Serrano heard crickets from his students, but 18 of them then dropped out of class. Nine students remained enrolled but did not take the final exam. And Serrano said the results proved him right; Three students received a zero, and the average score in the final was 48.6 percent — by far an all-time low, he said. Previously, the average final exam grade had never fallen below 65 percent. Only a few students performed as well as in the midterm.
In a follow-up message to students, Serrano announced that the midterm exam was invalid and the final exam would count for 80 percent of the students’ final grade. Any student who achieved a score of 40 percent or higher in the final exam received a passing grade – previously the minimum passing score was set at 50 percent or higher. A total of 19 students failed the course.
Table of Contents
“We cannot choose to become idiots”
In May, Serrano submitted the data shown above to Brown’s Standing Committee on Academic Code and received no response. After he went public with his story in late June, the committee, through its department head, asked Serrano to file individual complaints against any student suspected of cheating, including copies of their exams, he said.
“What they’re asking for is ridiculous… I think they’re planning on running them through an AI detection tool that’s known to give a lot of false positives and false negatives,” he said. “Your response, I have to tell you, is seen as appalling and inadequate by hundreds of people who have emailed me in support, including many Brown alumni.”
When asked about the university’s response to the fraud, Brown spokesman Brian Clark responded In Higher Ed that the process for investigating allegations of cheating is the same whether it involves one or more students.
“Brown treats any allegation of academic integrity with the utmost seriousness. Regarding this economics course, several Brown academic leaders have been in contact with the faculty member who raised concerns to provide details about how the allegations raised might be formally adjudicated. To date, the faculty member has not provided the Academic Code Standing Committee with the necessary details to pursue this path to resolution,” Clark wrote.
Serrano will meet with his dean on Wednesday to further discuss the incident, he said.
Adjudicating cheating allegations when so many students are suspected of cheating is tough on faculty, said Tricia Bertram Gallant, director of the Academic Integrity Office and the Triton Testing Center at the University of California, San Diego.
“They are not being rewarded for dealing with a 60-person fraud case,” she said. “They have no incentive to actually prevent fraud or report it when it occurs… Their time is not compensated. If the professor is an instructor and he gets paid the first day of class and stops on the last day of class, when are they supposed to do that? Or graduate students teaching in the summer?”
At the same time, institutions can’t sacrifice equity for efficiency, she said, but there are some steps colleges can take to find a middle ground. Bertram Gallant recommends that academic integrity committees allow students to take responsibility for cheating via email, rather than just during an in-person meeting, and that professors have the ability to file complaints against multiple students at once. Ultimately, institutions that care most about maintaining academic integrity will dedicate staff and resources to the matter, she said.
“Edgar Schein, an organizational theorist, said that leaders show how valued they are less by what they say and more by what they spend their money on,” she said. “The fact that most universities – especially the rich ones – do not have paid academic integrity experts is symbolic.”
As the cheating scandal unfolded in Serrano’s classroom, a Brown Committee on Generative AI in Teaching and Learning examined how the technology was being used at the university and made recommendations for how Brown can adapt and respond to AI developments. The inaugural report was released on Tuesday.
According to feedback from 105 faculty members, three-quarters of Brown professors said they were concerned about students using AI to cheat. The same share of respondents said the same thing in a 2025 national survey by the American Association of Colleges and Universities. As part of a series of mid-term recommendations, the committee encouraged the university to amend the College Academic Code and the Graduate Student Code “to reflect GenAI realities and protect against abuse.”
“Academic codes should address integrity, but also potential harm to students’ learning and personal development,” the report says. “Beyond the campus baseline regarding GenAI use, it is important that clear limits on misuse are enforced. Therefore, academic codes should explicitly address higher-level questions such as: Is something in “your own words” when GenAI provides grammatical improvements? Are the thoughts “your own” if you formed them through a brainstorming exercise with GenAI in the same way you would engage a colleague in a conversation? How do we apply the concepts of knowledge production, original research, citation and attribution in context? from GenAI?”
The committee also suggests that faculty “deemphasize punishment” and avoid extremely restrictive rules for the use of generative AI.
“There is no way to verify with 100 percent accuracy whether GenAI has been deployed, and standards are likely to change in the coming years,” the report said. “Moving the conversation beyond punishment also allows for the open dialogue that will be necessary as Brown navigates this moment.”
As colleges and universities grapple with AI, fraud needs to be taken seriously, Serrano said. “We cannot afford to have a society in which a significant proportion of our brightest young minds believe that cheating is OK,” he said. “This leads to a decaying society, a failed society… We cannot choose to become idiots.”
https://www.insidehighered.com/news/faculty/learning-assessment/2026/07/08/brown-professor-suspects-most-his-class-used-ai-cheat
