A team led by Dr. Patrick Traynor and FICS Research director Dr. Kevin Butler recently concluded the largest study on audio deepfakes to date, challenging 1,200 humans to identify real audio messages from digital fakes. “We found humans weren’t perfect, but they were better than random guessing. They had some intuition behind their responses. That’s why we wanted to go into the deep dive — why are they making those decisions and what are they keying in on,” said lead student author Kevin Warren, a FICS Research Ph.D. student in the Department of Computer & Information Science and Engineering. The study analyzed how well humans classify deepfake samples, why they make their classification decisions, and how their performance compares to that of machine learning (ML) detectors, noted the authors of the paper, “Better Be Computer or l’m Dumb: A Large-Scale Evaluation of Humans as Audio Deepfake Detectors,” which received a Distinguished Paper Award at the 2024 ACM Conference on Computer and Communication Security.

This originally appeared in the UF News article published on November 15th. The full article can be read here.