A dispute over an alleged AI-assisted exam at Yale has expanded into a 13-count federal lawsuit, according to Ars Technica. The case centers on Thierry Rignol, an Executive MBA student who says Yale wrongly punished him after his final exam was flagged by GPTZero.
Rignol says he paid $208,500 in tuition, was on track to graduate first in his class and was suspended for a year after the cheating allegation. He also received an F in the course Sourcing and Managing Funds.
The case illustrates a growing institutional problem: AI-detection tools can influence high-stakes decisions even though their reliability is contested. GPTZero and similar systems attempt to estimate whether text was generated by AI, but a detector score is not the same as direct proof of misconduct.
The lawsuit has been running since February 2025 and appears far from trial. Whatever the outcome, the dispute shows why schools need clear procedures, appeal paths and evidence standards before using AI detectors to discipline students.