The Noplag Database
The corpus built from submitted documents — what it stores, what a match against it reveals, and how to opt out.
When a report lists Noplag Database as a source, your text matched one or more documents previously submitted for checking. This page explains what that corpus is and how it treats your own submissions.
What it is
The Noplag Database is the corpus built from documents users submit for checking. Its purpose is protective: once your document has been checked, text copied from it will match in anyone else's later check. Your work defends itself by having been checked.
What a match reveals
A Noplag Database match is deliberately anonymous:
- The report says your text matches earlier submissions and shows the match size — not who submitted the source document, and not its full text.
- Matching runs on stored fingerprints (the engine's compact numeric representation — see the glossary), which can't reconstruct the original document.
Both sides of the trade
Your submissions help protect everyone's originality, and everyone else's submissions help protect yours. Someone plagiarizing an unpublished essay that was checked last month gets caught — that's the point.
Your choices
- Opt out — contributing to the Noplag Database is on by default and can be turned off; see the Privacy Policy for the current mechanism and scope.
- Deletion — deleting your account removes your documents' Noplag Database fingerprints along with everything else; see Account and privacy.
- Documents are never used for AI training — the database exists for matching, nothing else.
Self-hosting note
The Noplag Database is a hosted-product corpus. A self-hosted engine matches only against the corpus you build yourself — nothing you submit to your own instance reaches Noplag.