Skip to content
Submit incident
AII-1000-27063
ReportedReported by fewer than three independent publishers, with no official finding.

Brown University students suspected of using AI to cheat on exam

Professor Roberto Serrano at Brown University reportedly detected widespread AI-assisted cheating on a March midterm exam, where 40 of 86 students scored perfect marks and the class average was 96. When the same content was tested in a proctored final exam, the average score dropped to 48, prompting Serrano to eliminate take-home exams from his course.

5 March 2026 Β· Roberto Serrano Β· Fraud, scams and manipulation
5 March 2026
AII-1000-270632 sources ↓
Quick facts
Category
Fraud, scams and manipulation
Country
United States
Source
Futurism
Date
5 March 2026
Drafted by
Raidu Incident Detection

What happened

On a March 5 midterm exam for an advanced mathematical economics course at Brown University, 40 out of 86 students scored a perfect 100, with an average score of 96 out of 100. Professor Roberto Serrano reported that some answers contained unusual passages that coincided with results obtained by running the questions through ChatGPT.

On the in-person final exam, which accounted for half the final grade, the average score dropped to 48 out of 100. Of 27 students who did not show up for the final exam, 22 had scored 100 on the midterm, according to Serrano.

Reported impact

Serrano stated that 'the empirical evidence of fraud is overwhelming' and decided to stop giving take-home exams as a result. The incident highlights broader concerns about students resorting to AI tools to score high grades while contradicting honor codes they swore to uphold.

Index analysis

The unusually high midterm scores followed by dramatic collapse on the proctored final exam suggested students used AI to generate answers on the take-home exam. The incident reflects growing tensions at Ivy League schools over AI-facilitated cheating, with Princeton recently ending a 133-year 'Honor Code' tradition involving professors leaving the room during exams.

Governance controls

Brown University's academic code reportedly lacked explicit rules on AI use at the time, leaving faculty to fight fraud alone. Some professors lamented becoming 'plagiarism cops' focused on rooting out AI-facilitated cheating instead of teaching.

Classification

Category
Fraud, scams and manipulation
Industry
Education and training
Country
United States
Harm status
Actual harm
Intent
Intentional
Timing
Unknown
Harm types
Reputational
Failure modes
Hallucination
AI systems
ChatGPT

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

Futurism
Brown University Professor Horrified to Discover Largest AI Cheating Scandal in Ivy League History
30 June 2026
Gadget Review
96% on the Midterm, 48% on the Final: How a Professor Trapped 50 AI Cheaters Using Simple Math
30 June 2026