Friday, Jul 24, 2026 The claims desk. Receipts included. POWERED BY LENZ
IsThis

TECH

The Claim

As of 2026, AI-generated videos are realistic enough to fool the majority of viewers without the use of technical detection tools.

The Short Version

The strongest peer-reviewed evidence directly contradicts this claim. A large 2026 University of Florida controlled study published in PubMed found that humans correctly identified deepfake videos approximately two-thirds of the time — meaning most viewers are not fooled. Sources supporting the claim rely on qualitative assertions about realism or low-authority industry statistics with unclear provenance that contradict the gold-standard empirical findings. The claim overgeneralizes from specific high-quality deepfake scenarios to all AI-generated video.

Caveats

  • The most rigorous peer-reviewed study (University of Florida, 2026) found humans correctly classified deepfake videos ~67% of the time, directly refuting the 'majority fooled' threshold.
  • Industry statistics citing a '24.5% human detection rate' for high-quality deepfakes appear in multiple low-authority sources with no verifiable primary study, likely representing circular reporting.
  • Claims that AI video is 'virtually indistinguishable' describe technical realism in specific contexts but do not measure whether actual viewers are deceived — qualitative assessments of realism are not equivalent to empirical fooling rates.

The Receipts

  1. Is this real? Susceptibility to deepfakes in machines and humans

    PubMed

  2. Is this real? Susceptibility to deepfakes in machines and humans - PMC - NIH

    PMC - NIH

  3. Machines spot deepfake pictures better than humans, but ... - UF News

    UF News

  4. Seeing What Matters: Generalizable AI-generated Video Detection ...

    NVIDIA Research

  5. Seeing Is Not Believing: Realistic AI Videos Disrupt Confidence in Authentic Videos and Perceived Reality - MIT Media Lab

    MIT Media Lab

  6. Deepfakes leveled up in 2025: Here's what's coming next

    University at Buffalo

  7. FULL PAPER - ResearchOnline@JCU

    ResearchOnline@JCU

  8. Report: No Foolproof Method Exists for Detecting AI-Generated Media

    Campus Technology

  9. AI models spot deepfake images, but people catch fake videos

    Science News

  10. Deepfakes leveled up in 2025: Here's what's coming next - UBNow

    UBNow

+ 18 more sources — see the full list on Lenz

Filed Under

Artificial Intelligence Video GenerationDeepfake Detection

More Fact Checks