Astronomers Develop New Technique to Detect AI Deepfakes

Researchers use galaxy analysis tools to identify inconsistencies in light reflections within human eyes.

The method employs the Gini coefficient and CAS system, traditionally used in astronomy, to analyze eyeball reflections.
A study led by University of Hull researchers showed a 70% success rate in identifying deepfakes.
The technique was presented at the Royal Astronomical Society's National Astronomy Meeting.
While promising, the method is not foolproof and may evolve alongside AI advancements.
Deepfakes pose significant risks, including misinformation and privacy violations.