CoChat
Sign Up
Cited by 66Digital Media Forensic DetectionMisinformation and Its ImpactsDeception detection and forensic psychology

Human performance in detecting deepfakes: A systematic review and meta-analysis of 56 papers

Open access
Diel, Alexander, Lalgi, Tania, Schröter, Isabel Carolin, MacDorman, Karl F., Teufel, Martin, Bäuerle, Alexander·2024·Computers in Human Behavior Reports
DOI 10.1016/j.chbr.2024.100538
Open this paper in CoChat — free account

Abstract

Deepfakes are AI-generated media designed to look real, often with the intent to deceive. Deepfakes threaten public and personal safety by facilitating disinformation, propaganda, and identity theft. Though research has been conducted on human performance in deepfake detection, the results have not yet been synthesized. This systematic review and meta-analysis investigates human deepfake detection accuracy. Searches in PubMed, ScienceGov, JSTOR, Google Scholar, and paper references, conducted in June and October 2024, identified empirical studies measuring human detection of high-quality deepfakes. After pooling accuracy, odds-ratio, and sensitivity ( d' ) effect sizes ( k = 137 effects) from 56 papers involving 86,155 participants, we analyzed 1) overall deepfake detection performance, 2) performance across stimulus types (audio, image, text, and video), and 3) the effects of detection-improvement strategies. Overall deepfake detection rates ( sensitivity ) were not significantly above chance because 95% confidence intervals crossed 50%. Total deepfake detection accuracy was 55.54% (95% CI [48.87, 62.10], k = 67). For audio, accuracy was 62.08% [38.23, 83.18], k = 8; for images, 53.16% [42.12, 64.64], k = 18; for text, 52.00% [37.42, 65.88], k = 15; and for video, 57.31% [47.80, 66.57], k = 26. Odds ratios were 0.64 [0.52, 0.79], k = 62, indicating 39% detection accuracy, below chance (audio 45%, image 35%, text 40%, video 40%). Moreover, d' values show no significant differ…

Integrity signals

Registry

Crossref · 10.1016/j.chbr.2024.100538

Cites

115 of 115 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Diel and colleagues (2024) found total deepfake detection accuracy of 55.5% (audio: 62.1%, images: 53.2%, video: 57.3%), which is not significantly above chance level.

    “They found the total deepfake detection accuracy of 55.5% (audio: 62.1%, images: 53.2%, video: 57.3%), which is not significantly above the chance level.”

    From Deepfakes, Democracy, and the Ethics of Synthetic Media: A Synthesis of the SOLARIS Project (2026) · cites this paper as Diel et al., 2024

  • Diel and colleagues (2024) synthesized evidence on human ability to detect deepfakes of different modalities, finding 56 studies involving more than 86,000 participants.

    “Diel and colleagues (2024) synthesized the evidence on the human ability to detect deepfakes of different modalities, including audio, image, and video.They found 56 studies involving more than 86,000 participants that involved some kind of deepfake stimuli and detection performance measures (which varied between the studies).”

    From Deepfakes, Democracy, and the Ethics of Synthetic Media: A Synthesis of the SOLARIS Project (2026) · cites this paper as Diel et al., 2024

  • Human-centered methods often struggle to accurately identify AI-generated content.

    “Human-centered methods, which are dependent on individual judgments, often struggle to accurately identify AI-generated content (Groh et al. 2024; Somoray and Miller 2023; Diel et al. 2024; Bray, Johnson, and Kleinberg 2023).”

    From Characterizing AI-Generated Misinformation on Social Media (2025) · cites this paper as Diel et al. 2024

  • Analogous search strategies using the first 500 results of Google Scholar have been employed in prior meta-analyses.

    “Analogous search strategies using the first 500 results of Google Scholar have been employed in prior meta-analyses164.”

    From A scoping review on the mental health harms of LLM-based chatbots (2026) · cites this paper as [164]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Diel, A., Lalgi, T., Schröter, I. C., MacDorman, K. F., Teufel, M., & Bäuerle, A.
Affiliations
  • Essen University Hospital (DE)
  • Indiana University Indianapolis (US)
Corresponding author
Diel, Alexander
ORCID
Venue
Computers in Human Behavior Reports
Publisher
Elsevier BV
Published
2024-12
Volume
16
Pages
100538
ISSN
2451-9588 (ISSN-L)
DOI
10.1016/j.chbr.2024.100538
OpenAlex
W4404877539
Cited by
64
References
115
Citations by year
2026: 50 · 2025: 14
FWCI
2.26
Type
journal article
Language
en
Open access
Licence
Locations
  • Computers in Human Behavior Reports (publishedVersion, cc-by-nc-nd)
  • IUScholarWorks (Indiana University) (submittedVersion, cc-by-nc-nd)
  • Universitätsbibliographie, Universität Duisburg-Essen (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
  • Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
  • Social Sciences › Psychology › Social Psychology › Deception detection and forensic psychology
Keywords
deepfakes, deepfake detection, deepfake audio, disinformation, feedback training, media literacy, deception detection, media credibility

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review