Blind economics professor Roberto Serrano at Brown University, suspecting widespread AI cheating, cancelled the scheduled take‑home exam and required students to sit for an in‑person final. The abrupt change resulted in average scores falling about half of what the class had previously earned.
Serrano’s action reflects a broader anxiety among elite schools. A recent Princeton survey found nearly 30% of students admitted to using generative AI on at least one assessment, underscoring how competitive pressure can push even high‑achieving students toward shortcuts.
The incident may prompt other institutions to tighten exam protocols and reconsider the role of AI in coursework, as educators grapple with balancing technological tools against academic integrity.
