Provenance signals are pieces of information that help describe the origin, history and integrity of a digital asset.
In the C2PA ecosystem, provenance signals may come from Content Credentials, C2PA manifests, signatures, certificates, declared actions, ingredients, timestamps and trust-chain information.
These signals help users make a more informed decision about a piece of media. They do not automatically decide whether the content is true or false.
Examples of provenance signals
A C2PA verification tool may expose signals such as:
- whether Content Credentials are present;
- whether the C2PA Manifest is valid;
- whether the asset still matches the manifest;
- which tool or service signed the manifest;
- whether declared actions are available;
- whether the certificate chain is trusted;
- whether the media was created, edited, exported or recompressed by declared tools;
- whether ingredients or previous assets are referenced.
Provenance signals vs AI detection
Provenance verification and AI detection are different.
AI detection tries to guess whether a file was generated or altered by AI based on patterns or model-specific signals. C2PA provenance verification checks declared, signed and verifiable metadata when it exists.
A C2PA credential may include information about AI tools or actions, but ProvSeal does not claim to detect all AI-generated content. It verifies available provenance and integrity signals. Read more in C2PA vs AI detection.
Provenance signals in ProvSeal
ProvSeal helps users read provenance signals directly in Chrome.
The extension can verify local files, scan visible media on a page, show status badges and open a detailed report. This makes it easier to understand whether a file has C2PA evidence, whether that evidence is valid, and whether the trust chain is recognized — including cases marked Signed but untrusted.
Compare signed provenance with other signals in C2PA vs Watermarking and C2PA vs EXIF.