What FauxGuard does
A visitor can upload a JPG, PNG, or WebP image and receive an estimated AI-generation probability, a confidence band, readable provenance information, and plain-language reasons for the result. When an image contains camera metadata or a known generator signature, those clues are surfaced directly. When the file itself is inconclusive, a third-party visual detection model checks for generative and deepfake-related patterns.
The output is intentionally more detailed than a simple “real” or “fake” label. Image origin is often uncertain. Metadata can be removed by a messaging app, edited by a conversion tool, or changed deliberately. A camera-origin file can be heavily edited. An AI-generated image can be compressed until its original provenance is gone. FauxGuard presents the available evidence and the confidence level instead of hiding that uncertainty behind a final-sounding claim.
Who runs the site
FauxGuard is independently operated under the fauxguard.ai domain. Product decisions, editorial content, technical maintenance, and support are handled by the team behind FauxGuard. The public contact address is hello@fauxguard.ai.
FauxGuard is not a law-enforcement agency, forensic laboratory, identity-verification service, newsroom, or court. It does not claim institutional authority that it does not have. Questions about a result, a correction, privacy, or a commercial use case can be sent to the same contact address.
Who the detector is for
The tool is built for people who need a quick first-pass signal: journalists checking a circulated image, community moderators reviewing a suspicious profile or post, marketplace teams looking at disputed product evidence, researchers exploring generative-media patterns, and individuals who want a second opinion before trusting or sharing a photo.
It is also useful for learning what different image signals mean. A camera tag does not automatically make a file authentic. A missing camera tag does not automatically make it synthetic. A high visual score is a reason to look more closely, not a reason to accuse someone. FauxGuard is most useful when the result becomes one part of a documented review process that also considers the source, context, original file, account behavior, and applicable policy.
How we treat evidence
We favor explanations over certainty. Every result separates what was found in metadata from what a visual model inferred, and it includes a confidence band. The reasoning should help a person decide what question to ask next: Is there an original file available? Does a known generator signature appear? Was editing software named? Is the claim consistent with independent reporting or platform records?
We also avoid presenting a detection score as a standalone verdict. Automated image detection has limits. Models can be conservative on some synthetic images and overconfident on unusual real photos. Compression, filters, screenshots, and format conversion can remove important clues. A responsible workflow therefore keeps human judgment and a way to appeal or correct a mistaken decision.
What the service does not do
FauxGuard does not create a public image history, provide user accounts, publish uploaded files, or claim ownership of your images. It does not use the free detector to train a foundation model. It does not determine whether a statement is true, whether a person committed fraud, or whether an image would be accepted as evidence in a legal proceeding.
FauxGuard is designed for inspection and explanation, not for altering an image or disguising how it was made. More detail about data handling is available in the Privacy Policy, and the rules for using the detector are in the Terms of Service.
Contact and corrections
If an explanation is unclear, a page contains a factual error, or you represent an organization evaluating FauxGuard for a legitimate evaluation, contact hello@fauxguard.ai. Include the page URL and a concise description of the issue. For privacy or deletion questions, avoid sending sensitive images unless they are necessary to identify the request.
About FAQ
Who runs FauxGuard?
FauxGuard is independently operated under the fauxguard.ai domain. Product, editorial, and support decisions are handled by the FauxGuard team, which can be reached at hello@fauxguard.ai.
Does FauxGuard store my uploaded images?
FauxGuard does not maintain an account history or a gallery of uploads. An image is processed for the current request and may be sent to a third-party visual detection service when visual analysis is required. Read the Privacy Policy for the full explanation.
Is FauxGuard a forensic or legal service?
No. FauxGuard is an informational image-detection tool. Its result is not a legal finding, a forensic report, or proof that a particular person created an image.
How can I contact FauxGuard?
Send product, privacy, correction, or business questions to hello@fauxguard.ai. Please do not send sensitive images unless they are needed to explain a specific issue.