Super-Recognisers can Detect AI-hyperrealism

The AI revolution has produced synthetic faces that often appear more human than real photos of people. We tested whether individual differences in face recognition abilities explain variation in discriminating AI from human faces. Super-recognisers—people with exceptional ability to recognise human faces (N = 36)—outperformed a standard participant group by 15 percent, and by 7 percent compared to motivated controls (Cohen’s d = 0.55; N = 89). Individual difference analysis revealed that this pattern reflected an overall graded association between human face recognition and AI-face discrimination abilities. AI-discrimination ability was also associated with individual’s sensitivity to the ‘hyper-average’ appearance of AI-faces, with deep neural networks optimised for face identity processing confirming more central distribution of AI-faces in facespace. Super-recogniser’s correct interpretation of hyper-averageness as a cue to artificiality constitutes the first mechanistic link between evolved expertise in face processing and AI-face detection—and addresses a common misconception regarding the statistical structure of human facespace.