AIPhys.org Technologyabout 1 hour ago
Keele University AI Detects Fake Faces
2 min read
Researchers at Keele University developed FDIP-NET, a physics-inspired AI model that distinguishes genuine faces from deepfakes using color and brightness patterns. The tool achieved 98.654% accuracy on the DFRI dataset and 99.669% on RFF-140K. Published in Frontiers in Imaging in 2026, the study highlights the method's potential for combating identity theft and misinformation in digital media.
Level
Hype check
Tap to vote and see what everyone thinks.
