Provenance Is Not Deepfake Detection: What Each One Actually Does
By the Ingenious AI team · Published: November 14, 2026 · Reviewed: November 14, 2026 · 8 min read
The short answer
Detection examines a file of unknown origin and estimates whether it is synthetic; provenance records an image at creation and verifies copies against that record. ProvenanceGuard does the second and not the first: it cannot judge an unregistered stranger's image, and what it says about a registered one, hashes, fingerprints, timestamps, is measurement rather than estimation.
Asking a registry to detect deepfakes is asking a land registry to spot forged houses; it knows its own records perfectly and the rest of the world not at all.
The limits deserve the same prominence as the features. ProvenanceGuard does not detect deepfakes, does not judge whether an image is harmful, and does not verify that a creator's declared disclosure is true; it records declarations, binds them to fingerprints, and makes them checkable. Its signed assertion is its own XMP-embedded, HMAC-signed statement, explicitly not a C2PA manifest and not a certification by any standards body. Survival re-checks are local simulations of documented platform transforms, not actual uploads, and they are labelled that way in every result.
What is deepfake detection? analysing media for signs of synthesis or manipulation after the fact, producing a probability judgement about content whose origin is unknown, unlike provenance, which records origin at creation.
Why provenance versus detection matters
Used inside those limits, the record is strong. Used outside them, by claiming a seal proves authorship in law, or that a simulation proves a platform's behaviour, or that a registry match proves ownership, the same record becomes a liability, because the first competent challenge will find the overclaim. The teams that get the most from provenance tooling are the ones whose public language is as precise as the tool's own: sealed, registered, verifiable, with the dates and hashes to show it.
The registry is the part that survives distribution. At sealing, the asset's SHA-256, its sealed copy's SHA-256, and its perceptual fingerprints are stored against a public exhibit ID. Anyone with that ID, or with the file's hash, can open the public verification page or query the verification API and confirm that this exact file, or a near copy of it, was registered, when, and with what declared creator and disclosure. Verification pages are public by design: a record only a vendor can check would not settle disputes.
How the record is built
Two kinds of fingerprint do two kinds of work. The SHA-256 is exact: change one byte and it changes completely, which makes it perfect for proving a copy is bit-for-bit the registered file and useless for recognising a resized repost. The perceptual fingerprints, computed from the pixels rather than the bytes, are approximate on purpose: they stay close when a copy is resized, re-encoded or lightly edited, and drift as edits accumulate. A good registry stores both and is honest about which kind of match it is reporting.
The registry model also disciplines what a match means. A fingerprint match says this image is the registered image or a close descendant of it; it does not say the person showing it to you owns it, and it does not say the disclosure attached to it was correct. Those judgements stay human. What the registry removes is the argument about identity and timing: which file, registered when, carrying which declaration. Most real disputes shrink considerably once those three facts stop being contestable.
Where it meets the real world
Provenance is the record of where a piece of content came from and what has happened to it since: who or what created it, when it was registered, and which copies descend from the original. For an AI-generated image that record matters because the file itself is easy to copy, rename, crop and re-upload, and none of those actions leaves a reliable trace inside a filename. A provenance record kept outside the file, in a registry keyed to the file's fingerprints, survives exactly the handling that erases the signals embedded in the file.
The practical problem is distribution. An image is rarely published once, in one place, in its original form. It is uploaded to platforms that resize and re-encode it, forwarded by partners, saved and reposted by audiences, and archived by people who strip whatever context surrounded it. Each step is normal and usually innocent, yet together they separate the image from anything that said how it was made. Planning for that separation, rather than hoping it will not happen, is the whole discipline this guide is about.
Putting it into practice
No single layer does this job alone, and honest tooling says so. Embedded metadata can be stripped. A visible label can be cropped. A fingerprint can drift if an image is edited heavily. A registry entry proves registration, not authorship in a legal sense. Used together, though, the layers cover each other's failures: when the metadata is gone the fingerprint can still find the registry record, and when a copy is too altered to match, the labelled version and the evidence pack still tell the story.
ProvenanceGuard is built around those layers, and its wording is deliberately narrow. The free job inspects one JPEG or PNG, embeds a signed ProvenanceGuard assertion in XMP, produces a visibly labelled copy, registers the SHA-256 and perceptual fingerprints, and gives the asset a public verification page. The assertion is HMAC-signed and labelled accurately everywhere as ProvenanceGuard's own assertion; it is not a C2PA manifest and is never described as one.
Five actions to take
- Say 'not a registered asset' when that is the finding, and stop there
- Seal your own output so your half of the conversation is provable
- Seal the final export of the asset, with the creator and disclosure declared in your own accurate words
- Download and keep both copies the seal produces: the sealed copy as the record, the labelled copy for publication
- File the exhibit's verification URL beside the asset wherever your team keeps campaign or client records
A worked example: A comms team asked to pronounce on a viral image
The image was not in the registry, so the team said exactly that, checked what could be checked, and declined to call it real or fake; meanwhile its own campaign images, all sealed, could be answered for in one lookup each.
Notice what made the difference: not a better argument, but a record created before it was needed, carrying dates and hashes a stranger could check. That is the whole pattern this guide recommends, applied at the scale of one team and one deadline.
FAQ
Can the fingerprint tools help with unregistered images at all?
Inspection can report what an unregistered file contains, markers, XMP, EXIF, which is factual and sometimes useful. It cannot turn absence of a record into a verdict about the image, and the product does not present it as one.
Is ProvenanceGuard's assertion a C2PA manifest?
No. It is ProvenanceGuard's own signed assertion, embedded as XMP and HMAC-signed, and labelled that way everywhere. The inspection detects genuine C2PA markers when a file carries them, but the two records are never conflated.
What formats and sizes are supported?
JPEG and PNG, up to 25 MB per file. Those are the formats the inspection, XMP embedding and fingerprinting are built and tested against; other formats are refused rather than approximated.
What does the free tier actually do?
The complete job on an asset: byte-level inspection, a signed assertion embedded as XMP, a visibly labelled copy, SHA-256 and perceptual fingerprints registered, and a public verification page. Paid plans add survival re-checks, evidence packs and bulk verification.