How Turnitin’s "no repository" mode keeps your Hong Kong submission clean
Updated May 26, 2026 · 8 min read
If you study at HKU, CUHK, HKUST or PolyU, your work almost certainly passes through Turnitin before a human marks it. And one question causes more late-night panic than any other: if I get help with a draft, will Turnitin flag it as a duplicate when I upload it myself?
The short answer is no — provided the work is original and scanned correctly. This guide explains how Turnitin’s two checks work, what "no repository" mode means, and what genuinely original means in Hong Kong’s OBTL-driven system.
Two checks, one report
Students treat Turnitin as a single "plagiarism score". It is really two separate signals:
- Similarity. The percentage of your text matching sources in Turnitin’s database — journals, web pages, and crucially, papers other students have already submitted.
- AI writing indicator. A separate estimate of how much of your document reads as machine-generated, based on patterns in how sentences are built.
A paper can score 3% similarity and still be flagged as AI-written — or the reverse. They are not the same number.
What "no repository" mode actually does
When a paper is scanned through a standard student account, Turnitin saves a copy to its repository. That is why submitting the same text twice triggers a near-100% match the second time. Private institutional accounts can scan in "no repository" mode — the document is checked against existing sources but is not stored. Nothing is added to Google’s index or Turnitin’s central database.
The practical effect: a draft can be checked for originality without ever becoming a future "match" against itself. When you later upload through your university portal, there is no stored copy to collide with.
Why Hong Kong universities care
Academic integrity is taken seriously across the territory. Institutions run Turnitin on submitted coursework, and OBTL frameworks mean assessors are looking for evidence you met specific learning outcomes — not just that you produced text. A misconduct flag can mean a failed course or a formal hearing, and for students on a visa tied to study progress, the stakes are higher still.
What this means if you used a free AI tool
Pasting AI output into a submission is risky on three fronts: the AI indicator may flag it, the references the model invented will not survive a source check, and the writing ignores your actual rubric. The cleanup almost always costs more stress than starting properly.
What genuinely original work looks like
- Written from a blank document by a human specialist — not generated and lightly edited.
- Real, verifiable references in your required style — APA, Harvard, OSCOLA or Chicago.
- Mapped to your OBTL outcomes and the HKQF level of your course.
- Scanned cleanly, with the report delivered to you.
How we keep your work clean and verifiable
Every assignment we deliver is written by a native HK subject specialist from a blank document, then checked through Turnitin (similarity + AI) in no-repository mode before it reaches you. You receive the report alongside your file, so originality is something you can see. If any section flags during our internal check, we rewrite it before delivery — and we supply every piece as a model research guide to learn from and build on ethically.
Key takeaways
- Turnitin reports similarity and AI writing as two separate signals.
- No-repository mode checks a draft without storing it — so it won’t collide with your own later upload.
- HK universities run Turnitin under OBTL frameworks and treat breaches seriously.
- The safe path is original, properly referenced, human-written work — with the report attached.
Native HK scholars, editors and former university staff writing practical guidance for students across the territory. We write everything ourselves — including this.
