Brown Probes AI Cheating Claims After Scores Drop 40 Points on In-Person Final
Updated
Updated · Business Insider · Jul 9
Brown Probes AI Cheating Claims After Scores Drop 40 Points on In-Person Final
3 articles · Updated · Business Insider · Jul 9
Summary
Brown's academic code committee received Roberto Serrano's allegations on July 8 after students who scored in the high 90s on a take-home midterm fell into the 50s on an in-person final.
The welfare economics professor suspected AI-assisted cheating because the at-home midterm produced unusually strong results, while many top scorers dropped the class once he announced the final would be held in person.
Brown said it treats every academic-integrity allegation seriously and will handle the case under committee procedures; Serrano said the score gap alone does not definitively prove mass cheating.
The episode, which spread widely online and drew hundreds of emails to Serrano, has intensified debate over whether AI is making cheating nearly costless for students.
Serrano said he will stop giving take-home exams and remove homework from grades, calling the case a wake-up call for professors to rethink AI policies.
When AI can pass university exams, what is the true value of a modern degree?
If take-home exams are dead, must we redesign education from the ground up?
50+ Students Caught in Brown University AI Cheating Scandal: How Generative AI Sparked an Academic Integrity Crisis in Spring 2026
Overview
In Spring 2026, Brown University faced a major academic integrity crisis due to the widespread use of artificial intelligence tools. This led to significant challenges for instructors, who found AI cheating to be a serious problem across departments. As a result, traditional assessment methods were reevaluated, especially after incidents in Professor Roberto Serrano’s course. Professor Serrano responded by ending take-home exams and removing homework from grading, highlighting how easily generative AI can enable dishonesty. These decisions show the growing difficulty in accurately assessing student learning in an environment where AI use is common.