A new scandal at Brown University has shaken the academic world. Professor Roberto Serrano, an economics professor, decided to administer an in-person final exam after suspecting widespread use of artificial intelligence by students. The result was startling: grades plummeted by 50% compared to previous years when exams were online. This data suggests that a large portion of students were using generative AI tools, such as ChatGPT, to complete assignments and pass remote assessments.
Professor Serrano's suspicion and the in-person test
Roberto Serrano, a blind professor at the Ivy League institution, had noticed a discrepancy between online performance and in-class work. To verify, he imposed a traditional written exam in person, without internet access. The results confirmed his fears: students scored on average half the points compared to earlier tests. Such a drastic drop cannot be attributed to random factors but clearly indicates systematic use of AI to cheat.
Sponsored Protocol
Implications for academic integrity
This episode highlights a growing challenge for universities. According to a recent Princeton survey, 29.9% of students admitted to using AI on at least one exam or assignment. The Brown case, however, provides concrete evidence of the phenomenon's scale. Professor Serrano stated that "we cannot choose to become idiots," emphasizing how AI abuse undermines real learning. Universities are now seeking strategies to combat these academic frauds, from anti-plagiarism software to reintroducing in-person exams.
The broader context of AI in education
Generative AI offers enormous educational potential, but its misuse risks producing a generation of students unable to think critically. Tools such as those offered by OpenAI, discussed in a related article on OpenAI offers shares to Americans, are becoming increasingly accessible, making it easier for students to shortcut their studies. Similarly, companies like Meta are integrating AI into their products, as described in Meta defaults to using Instagram photos for AI. An urgent debate on the rules to adopt is needed.
Sponsored Protocol
Reactions and possible solutions
Brown University has launched an internal investigation, while other Ivy League schools are reviewing their policies. Some propose integrating AI as a learning tool rather than banning it; others insist on stricter controls. Professor Serrano, interviewed by Ars Technica, stated that AI misuse leads to "a failed society." For further information, consult the original article on Ars Technica.
