How DetectFake analyzes media—and where detection stops

1. The submitted image is normalized for visual analysis without changing the original file.
2. For video, the browser samples evenly spaced, downscaled frames; the original video remains on the device.
3. Each image or frame is evaluated for visual patterns associated with generation and manipulation.
4. Detected signals are weighted together into a probabilistic class and confidence score.
5. The interface presents the verdict, uncertainty, summary, and individual signals for human review.

Confidence describes how strongly the available visual evidence supports the reported class. It is not the probability that a named person created a fake, and it does not identify a specific generator or editing workflow. Similar visual artifacts may have several causes.

Results near the middle should be treated as inconclusive. Even a high-confidence result should be checked against provenance, context, and independent evidence before it is used for a consequential decision.

• Social-media compression, screenshots, and repeated re-encoding can erase forensic traces.
• Filters, denoising, sharpening, unusual lighting, and computational photography can resemble synthetic artifacts.
• Small images and short videos provide less evidence to examine.
• New generation methods may produce patterns that differ from previously observed media.
• A composite may contain both authentic and generated regions, making a single label incomplete.

For video, frame extraction and downscaling happen locally in the browser, and only sampled frames are submitted for analysis. The original video is not uploaded. Scan history is kept locally in the user’s browser. See the privacy notice for account and service data details.

DetectFake is suitable for preliminary review, moderation triage, scam checks, and deciding whether deeper investigation is warranted. It is not a substitute for a qualified forensic examination, chain-of-custody procedures, or independent verification in legal, employment, financial, medical, or safety-critical settings.

For a repeatable verification process, read our guide on detecting AI-generated images.