- In a large study, AI programs were highly accurate at detecting AI-generated faces, known as deepfakes
- Human participants were unable to identify fake still photos, but performed computer at spotting fake videos
- Researchers say that sophisticated deepfakes require more vigilance to detect
Artificial intelligence may be better than people at spotting fake faces in photos — but humans still have the upper hand when those fakes start moving.
In a large recent study, psychologists and computer scientists at the University of Florida found that AI programs were up to 97% accurate at detecting pictures of deepfake faces. Participants in the study performed no better than chance.
However, the algorithms’ performance declined sharply when it came to detecting deepfake videos. In those tests, programs performed at chance levels, while humans correctly identified real and fake videos about two-thirds of the time. Human participants appeared to pick up on subtle inconsistencies in movement, facial expressions and timing — cues the algorithms struggled to interpret.