Humans reach 64.8% accuracy detecting synthetic legal evidence images overall but drop to chance levels on top generators, while MLLMs achieve 100% specificity yet only 5.9% detection on the hardest synthetics, with uncorrelated error patterns.
Disrupting and Preventing Deepfake Abuse: Exploring Criminal Law Responses to AI-Facilitated Abuse
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it