UC Berkeley Math Professor Criticizes Students' Skills in AI-Assisted Op-Ed
UC Berkeley math professor Zvezdelina Stankova admitted to using AI to help edit a 2,000-word op-ed that criticized her students' poor math skills. The AI usage was discovered by a student journalist using the AI detection tool Pangram, which flagged 33% of the article as AI-assisted. This incident highlights the growing ethical debate surrounding AI usage by academics, especially when criticizing students' academic readiness. It also intersects with the ongoing controversy over whether the University of California system should reinstate standardized testing like the SAT and ACT. Stankova defended her actions, stating the article was the result of hundreds of hours of work, including 80 hours of her own, and that she also used AI to edit a June open letter signed by over 3,000 faculty members. While Pangram claims a 99.66% accuracy rate, members of UC's AI council noted that most AI detection tools remain highly unreliable.
## BACKGROUND
In 2020, the University of California Board of Regents voted to phase out standardized testing requirements (SAT/ACT) for admissions, a move that critics argue has led to a decline in student preparedness for rigorous college courses. Pangram is an AI detection software developed by Pangram Labs (formerly Checkfor.ai) designed to identify text generated by large language models.