Sometimes—but people are not reliable at spotting manipulated media by sight or sound alone. A 2024 meta-analysis found that, across the deepfake stimuli it examined, human detection was not significantly better than chance. The important gap is between feeling sure and having evidence: confidence is a subjective judgment, not an authenticity check.
How accurate are people at spotting deepfakes?
A 2024 systematic review and meta-analysis by Alexander Diel and co-authors combined 137 effects from 56 papers, involving 86,155 participants. For deepfake stimuli, pooled accuracy was 55.54% (95% confidence interval: 48.87%–62.10%; 67 effects). Because that interval includes 50%, the result does not establish reliable above-chance detection. It is an average across studies, not a prediction that every person will get roughly half of every test right.
As an Amazon Associate I earn from qualifying purchases.
“Near chance” means performance, at the pooled level, was not reliably distinguishable from random guessing under the conditions studied. It does not mean nobody ever notices a fake. Results can vary with the task, sample, media, and study design. The figures describe identifying synthetic or manipulated content, not determining whether a post’s underlying claim is true.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Does the type of media make a difference?
The same review reported different pooled point estimates for deepfake stimuli by format. The confidence intervals for these estimates crossed chance, so the figures do not show that any one format is reliably easy to identify.
#1 Best Overall
| Media type | Pooled accuracy | How to interpret it |
|---|---|---|
| Audio | 62.08% | Confidence interval crossed 50%; not evidence of dependable detection. |
| Images | 53.16% | Confidence interval crossed 50%; not evidence of dependable detection. |
| Text | 52.00% | Confidence interval crossed 50%; not evidence of dependable detection. |
| Video | 57.31% | Confidence interval crossed 50%; not evidence of dependable detection. |
These values are pooled estimates from Diel and co-authors’ 2024 review, not head-to-head results from one uniform test. Small differences between them should not be treated as a ranking of which media people can spot best.
Why can confidence outpace accuracy?
People often make authenticity judgments from cues that seem persuasive—an unnatural expression, odd audio, or a detail that does not fit expectations. But a cue can be misleading, and plausible-looking media can still be synthetic. A strong feeling that something is genuine or fake does not itself establish how it was made.
Chen and Goh’s 2025 study examines how confidence and individual traits relate to human deepfake-video detection accuracy and calls for further investigation into miscalibrated confidence. Its findings should not be reduced to a claim that confidence is always unrelated to accuracy. The practical distinction is simpler: confidence is a report of certainty, while accuracy is whether the judgment was correct.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Can training or a detector solve the problem?
Tested strategies can help, but the evidence is bounded
Diel and co-authors found pooled accuracy of 65.14% after detection-improvement strategies (95% confidence interval: 55.21%–74.46%; 15 effects). The tested approaches included feedback training, AI support, and deepfake caricaturization. This supports the possibility of improvement in study settings; it does not show that a generic checklist or any particular training program will authenticate arbitrary online media or remain effective against new techniques.
Automated scores are not verdicts
A detector can provide another signal, but its output should not be treated as proof. Wang and co-authors’ 2024 survey of deepfake detection identifies transferability, interpretability, and robustness as reliability challenges, and notes a lack of reliable evidence for some real-world uses. A score produced on one file or under one set of conditions does not by itself settle authenticity.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What should you do when a post matters?
When the consequences of sharing or acting on a clip are significant, use corroboration and context rather than relying on your impression or a detector score alone. This is a prudent verification practice, not a guaranteed deepfake-detection recipe.
Rank #4
- Look for the original post or source and check whether it provides context about when, where, and by whom the media was made.
- Check whether independent, credible sources corroborate the event or claim. Authentic media can still be presented with a false caption or misleading context.
- Compare the post with other available accounts or footage, while remembering that repetition across accounts is not necessarily independent confirmation.
- If a detector is available, treat its result as one clue. Consider the tool’s stated scope and limitations, and do not convert a score into a definitive real-or-fake label.
- If the evidence remains uncertain, avoid presenting the media as verified or forwarding it as fact.
The central distinction is between spotting signs and establishing provenance. Human judgment can raise questions; corroborating context is what helps answer them.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallQuick Recap
Best Value
- Book - 1, 000 books to read before you die: a life-changing list (1000 before you die)
- Language: english
- Binding: hardcover
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
