iProov deepfake blindspot consumer study (industry report)
Source record
iproov-2025-consumer-study· 2025 · Industry report / press release
Evidence tier: E (industry) · Verification: unverified (compiler confidence: medium)
Bottom line: Vendor-reported: only ≈0.1% of participants classified every item correctly; older adults performed worse; confidence was high regardless of accuracy.
Citation
iProov (2025). iProov deepfake blindspot consumer study (industry report). Industry report / press release.
Verification entry (identifiers, resolvers, claims with source locations): /data/citations.json#iproov-2025-consumer-study
Study design
| Field | Value |
|---|---|
| Design | Online consumer test with mixed real and deepfake images and videos |
| Modality | image-face, video |
| Stimuli | Vendor-selected |
| Participants (N) | 2000 |
| N detail | UK and US consumers (approximate) |
| Topics | human-detection, overconfidence |
Results
| Metric | Value | Note |
|---|---|---|
| All items correct | ≈0.1% of participants | all-correct is a strict criterion, not mean accuracy |
Moderators
- Older adults performed worse
Limitations
- Not peer reviewed
- Vendor has commercial interest
- Methods not fully disclosed
Cited in
- MR-0001: Human Detection of Deepfakes: Accuracy, Bias, Confidence, and Training
Verification
| Field | Value |
|---|---|
| Status | unverified |
| Compiler confidence | medium |
| Checked on | — |
| Via | compiler knowledge |
Values on this page have not all been checked against the full text. Confirm them against the primary source before high-stakes use.