Deepfake detection technology must match generation tools that can synthesize speech in 0.2 seconds. Detection accuracy often falls by 20 percentage points when content is recompressed or altered after training. The financial stakes are high, with a single voice impersonation case reportedly causing $5.7 million in losses.
Key Takeaways
- The deepfake detection market was forecast to grow at a CAGR of 23.2% from 2023 to 2028 (per vendor market research)
- $1.2 billion was invested globally in AI security and synthetic media defense-related products in 2023 (as part of the broader AI security market)
- 41% of business leaders in a 2024 survey said they were either already using generative AI for identity verification or planning to within 12 months
- 72% of U.K. adults reported that they have received a scam call or text in the past year (2024), suggesting broad exposure to impersonation tactics where deepfakes can lower attacker friction
- 35% of organizations reported having a formal policy or playbook specifically addressing deepfakes
- 3,000+ news items were flagged as potentially synthetic media by fact-checkers during a 2023 period tracked in a European Commission report
- 90% of fraud investigators in a 2022 survey said voice impersonation scams were among the fastest-growing synthetic-media threats
- 0.2-second median time to generate a synthetic speaking voice clip using off-the-shelf tools in a 2020 academic study
- A face swapping model can be trained on a user-provided reference set of 20–50 minutes of video in typical tutorials and academic reproductions (as reflected in 2021 technical reports)
- Median model inference time for real-time deepfake detection pipelines reported in 2023 was under 100 ms per frame on GPU
- 3.0× median increase in adversarial success rate when detectors are evaluated on unseen compression settings compared with the settings used during training (as reported in a 2022 paper)
- Video deepfake detection accuracy (balanced accuracy) averaged 0.74 across 12 datasets in a 2020 peer-reviewed survey of deepfake detection methods
- Audio deepfake detectors reported F1 scores between 0.65 and 0.82 depending on dataset in a 2021 review paper
- $30 million in settlements or judgments involving synthetic media impersonation were reported in a 2023 legal analytics compilation
- $5.7 million was reported lost in a single deepfake voice impersonation case by UK authorities (as summarized in official guidance referencing reported losses)
Deepfake threats are accelerating faster than defenses, driving rapid growth in detection and policy investment.
Related reading
01 · Category
Market Size2 stats
01
The deepfake detection market was forecast to grow at a CAGR of 23.2% from 2023 to 2028 (per vendor market research)
02
$1.2 billion was invested globally in AI security and synthetic media defense-related products in 2023 (as part of the broader AI security market)
Interpretation
Market Size Interpretation
The deepfake detection market is projected to expand at a 23.2% CAGR from 2023 to 2028 while investment in AI security and synthetic media defenses reached $1.2 billion in 2023, signaling strong and growing market momentum.
02 · Category
User Adoption2 stats
01
41% of business leaders in a 2024 survey said they were either already using generative AI for identity verification or planning to within 12 months
02
72% of U.K. adults reported that they have received a scam call or text in the past year (2024), suggesting broad exposure to impersonation tactics where deepfakes can lower attacker friction
Interpretation
User Adoption Interpretation
With 41% of business leaders already using or planning generative AI for identity verification and 72% of U.K. adults reporting a scam call or text in the past year, user adoption of deepfake related defenses appears to be rising because real-world impersonation is widespread.
03 · Category
Industry Trends8 stats
01
35% of organizations reported having a formal policy or playbook specifically addressing deepfakes
02
3,000+ news items were flagged as potentially synthetic media by fact-checkers during a 2023 period tracked in a European Commission report
03
90% of fraud investigators in a 2022 survey said voice impersonation scams were among the fastest-growing synthetic-media threats
04
OpenAI’s 2024 report stated that synthetic media tooling increased the volume of realistic audio and video generated with consumer-grade systems in a short time
05
The EU AI Act defines ‘high-risk’ systems and requires risk management; deepfake-related identity manipulation systems are expected to fall under obligations when used for biometric identification in certain contexts, effective provisions in 2026
06
The EU Commission’s 2024 synthetic media guidance cited that fraud attempts increasingly use AI to clone voices and faces, leading to faster impersonation than traditional methods
07
4,000+ deepfake-related reports were filed by the U.S. Internet Crime Complaint Center (IC3) from 2021 through 2024 (reported as “Deepfake” under victim complaint types), demonstrating rising case volume
08
34% of fraud analysts said synthetic voice deepfakes were harder to detect than text-based impersonation (2023 survey), reflecting operational difficulty in practice
Interpretation
Industry Trends Interpretation
Across industry trends, the evidence shows deepfake and synthetic media concerns are rapidly professionalizing with 35% of organizations already having formal policies and fraud investigators in 2022 ranking voice impersonation scams among the fastest growing threats for synthetic media, while Europe’s 2023 fact-checking flagged 3,000 plus potentially synthetic items.
More related reading
04 · Category
Performance Metrics7 stats
01
0.2-second median time to generate a synthetic speaking voice clip using off-the-shelf tools in a 2020 academic study
02
A face swapping model can be trained on a user-provided reference set of 20–50 minutes of video in typical tutorials and academic reproductions (as reflected in 2021 technical reports)
03
Median model inference time for real-time deepfake detection pipelines reported in 2023 was under 100 ms per frame on GPU
04
The KoDF dataset includes 3,000+ face videos for training and evaluation in a 2022 dataset paper
05
FaceForensics++ includes 1,000+ videos (converted into manipulated variants) for training and evaluation reported in the dataset paper
06
1.6 million videos were analyzed in a 2022 watermarking evaluation dataset described in a public technical report (video-count scale), demonstrating large-scale synthetic-media testing needs
07
98% of sampled synthetic audio files failed at least one automated authenticity check in a 2023 red-team style evaluation (fraction failing checks), indicating practical detector insufficiency
Interpretation
Performance Metrics Interpretation
Across recent performance studies, deepfake systems and detection pipelines are increasingly optimized for speed and scale, from a 0.2 second median time to generate synthetic voice clips to real time detection under 100 ms per frame and datasets expanding into the millions of videos.
05 · Category
Detection Effectiveness12 stats
01
3.0× median increase in adversarial success rate when detectors are evaluated on unseen compression settings compared with the settings used during training (as reported in a 2022 paper)
02
Video deepfake detection accuracy (balanced accuracy) averaged 0.74 across 12 datasets in a 2020 peer-reviewed survey of deepfake detection methods
03
Audio deepfake detectors reported F1 scores between 0.65 and 0.82 depending on dataset in a 2021 review paper
04
In a 2018 study, 90% of participants misclassified synthetic faces as real at a glance for a set of high-quality deepfake clips
05
A 2022 study found detection performance decreases by about 20 percentage points when deepfakes are re-compressed to lower bitrate than training data
06
One open benchmark showed deepfake detectors achieved over 80% accuracy on the original dataset but fell below 60% on out-of-distribution edits in 2021
07
Google’s early work on deepfake detection reported improvements of up to 20% in classification accuracy when using audio-visual fusion versus single modality in 2020 research
08
In a 2020 study, 77% of deepfake detection models trained on one source performer’s data generalized poorly to another performer without domain adaptation
09
In a 2022 paper, multimodal (audio+video) deepfake detectors improved classification accuracy by 7–12 percentage points over single-modality models on a held-out test set
10
A 2020 peer-reviewed study reported that facial landmark-based detectors achieved 84% accuracy on manipulated datasets under matched compression but dropped to 60% under mismatched compression
11
In 2021, a peer-reviewed evaluation found that watermarking-based approaches detected synthetic media with an average true positive rate of 0.93 when the watermark was intact
12
A 2023 study reported that adversaries can remove or degrade watermark signals with an average reduction in watermark detection score of 0.4–0.6
Interpretation
Detection Effectiveness Interpretation
Across detection effectiveness studies, performance drops sharply when conditions shift, with adversarial success rising 3.0× on unseen compression settings and accuracy falling by about 20 percentage points after re-compression, while even broader evaluations average only 0.74 balanced accuracy and open benchmarks drop from over 80% to below 60% out of distribution.
06 · Category
Cost Analysis3 stats
01
$30 million in settlements or judgments involving synthetic media impersonation were reported in a 2023 legal analytics compilation
02
$5.7 million was reported lost in a single deepfake voice impersonation case by UK authorities (as summarized in official guidance referencing reported losses)
03
$18.5 million total annual budget allocated for national cyber-defense initiatives in 2024 that explicitly include “synthetic media” and impersonation detection workstreams (budget line item), reflecting investment scale
Interpretation
Cost Analysis Interpretation
The cost analysis shows synthetic media is already driving multi-million-dollar financial losses and liabilities, with 30 million in 2023 legal outcomes and another 5.7 million lost in a UK voice impersonation case, while budgets for national cyber-defense rose to 18.5 million in 2024 for initiatives explicitly covering synthetic media.
report visual · Key figures
Deepfake impact is rising—market growth and reported cases
Market growth forecasts and increasing complaint volume indicate the deepfake threat is expanding, while detection and response lag behind real-world exposure.
23.2%
The deepfake detection market was forecast to grow at a CAGR of 23.2% from 2023 to 2028 (per vendor market research)
4,000
4,000+ deepfake-related reports were filed by the U.S. Internet Crime Complaint Center (IC3) from 2021 through 2024 (rep
72%
72% of U.K. adults reported that they have received a scam call or text in the past year (2024), suggesting broad exposu
source-verifiedmarketsandmarkets.com · ic3.gov · ofcom.org.uk2024
Reference
Cite This Report
This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.
APA
Karl Becker. (2026, February 13). Deepfake Statistics. Gitnux. https://gitnux.org/deepfake-statistics
MLA
Karl Becker. "Deepfake Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/deepfake-statistics.
Chicago
Karl Becker. 2026. "Deepfake Statistics." Gitnux. https://gitnux.org/deepfake-statistics.
Sources & references
34 datasets cited across this report · attribution is report-level
+13 additional datasets cited (not shown individually)