NYU researchers develop new real-time Deepfake detection method
Associate Professor Chinmay Hegde is leading research to combat real-time deepfakes through challenge-response systems. His team developed eight visual and 22 audio challenges designed to expose AI-generated content. The tests include actions like hand gestures, facial expressions, and whispering — simple for humans but difficult for AI to replicate convincingly. "As AI gets better at creating fakes, we need to get better at detecting them," said Hegde.