iThenticate

iThenticate is a text-similarity-checking service built by Turnitin and marketed specifically to publishers, journal editorial offices, funding agencies, and research institutions — as distinct from Turnitin’s own education-market product, which is licensed to universities primarily to screen student coursework and thesis submissions. Both run on the same underlying detection engine; the difference is audience and licensing channel, not technology.

What iThenticate Actually Does

iThenticate compares an uploaded document — a manuscript, grant proposal, thesis, or dissertation — against a large reference corpus of current and archived web content, previously submitted documents, and licensed academic publications, journals, and periodicals. The output is a similarity report: an overall similarity index (the percentage of the submitted text with a detectable match somewhere in the corpus) plus a source-by-source breakdown showing which specific passages matched which specific sources. For a detailed breakdown of how that report is structured and interpreted, see originality report.

iThenticate vs. Turnitin vs. Crossref Similarity Check

These are not three competing products — they are the same detection engine offered through three different channels. Turnitin is licensed to universities for coursework and thesis screening, typically integrated into the institution’s learning-management system. iThenticate is licensed directly to publishers, journal editorial offices, and research institutions for manuscript and proposal screening, independent of any classroom system. Crossref Similarity Check is a service Crossref operates on top of the iThenticate engine, giving Crossref member publishers integrated similarity-checking access — often built directly into their editorial-management system — without licensing iThenticate separately. For the full comparison across all three, see the CASRAI guide anti-plagiarism software: how Turnitin, iThenticate, and Crossref Similarity Check work.

Where iThenticate Fits in Editorial and Research-Administration Workflows

In scholarly publishing, an iThenticate check is typically run as an early screening step — before or during initial editorial review — with the resulting report treated as one input the handling editor weighs alongside the manuscript itself, not an automatic accept or reject rule. Some funding agencies, including the U.S. National Science Foundation, have also used iThenticate to screen incoming grant proposals for text overlap as part of proposal intake. In both settings, a high similarity index does not by itself establish plagiarism: proper quotation, a shared reference list, standard methodological language, or a match against the authors’ own prior published work (see self-plagiarism) can all raise the percentage without indicating misconduct. Turnitin’s own guidance for the underlying similarity technology is explicit that the tool does not determine whether plagiarism has occurred; it supports, rather than replaces, a human reviewer’s judgment. Where a report raises a substantive concern beyond a routine false positive, it typically escalates into the publisher’s or institution’s research misconduct process rather than being resolved by the similarity score alone.

What iThenticate Does Not Do

iThenticate matches text strings against its reference corpus; it does not evaluate images, figures, underlying data, or code, and it cannot detect content that was never indexed into its corpus — paywalled sources the vendor has no license for, non-digitized material, or heavily reworded (rather than copied) passages can all go unmatched. AI-generated-text detection is a separate, distinct capability from similarity checking, even where offered by the same vendor, and should not be treated as an extension of plagiarism detection with the same evidentiary reliability.

Related Terms

• Originality report
• Plagiarism
• Self-plagiarism
• Research misconduct
• Anti-plagiarism software (guide)

Frequently Asked Questions

How does iThenticate work?

iThenticate compares an uploaded document — a manuscript, grant proposal, thesis, or dissertation — against a large reference corpus of current and archived web content, previously submitted documents, and licensed academic publications. It returns a similarity report consisting of an overall similarity index, the percentage of text with a detectable match somewhere in the corpus, plus a source-by-source breakdown showing which passages matched which sources.

What is iThenticate used for?

In scholarly publishing, iThenticate is typically run as an early screening step before or during initial editorial review, with the report treated as one input the handling editor weighs alongside the manuscript itself rather than an automatic accept-or-reject rule. Some funding agencies, including the U.S. National Science Foundation, have also used it to screen incoming grant proposals for text overlap during proposal intake.

Is iThenticate the same as Turnitin?

They run on the same underlying detection engine but are licensed through different channels to different audiences. Turnitin is licensed to universities, typically through the institution’s learning-management system, to screen student coursework and thesis submissions, while iThenticate is licensed directly to publishers, journal editorial offices, funding agencies, and research institutions for manuscript and proposal screening. See the CASRAI guide on anti-plagiarism software for the full comparison.

What is Crossref Similarity Check, and how does it relate to iThenticate?

Crossref Similarity Check is a service Crossref operates on top of the iThenticate engine, giving Crossref member publishers integrated similarity-checking access, often built directly into their editorial-management system, without requiring them to license iThenticate separately. It is not a competing product — it is the same detection engine offered through a third channel, alongside Turnitin and direct iThenticate licensing.

Does a high iThenticate similarity score mean a manuscript was plagiarized?

Not by itself. Proper quotation, a shared reference list, standard methodological language, or a match against the authors’ own prior published work can all raise the similarity index without indicating misconduct. Turnitin’s own guidance for the underlying technology states that the tool does not determine whether plagiarism occurred; it supports, rather than replaces, a human reviewer’s judgment, and substantive concerns typically escalate into the publisher’s or institution’s research misconduct process rather than being resolved by score alone.

Can iThenticate detect AI-generated text, images, or data manipulation?

No. iThenticate matches text strings against its reference corpus; it does not evaluate images, figures, underlying data, or code, and it cannot detect content that was never indexed into its corpus. AI-generated-text detection is a separate, distinct capability from similarity checking, even when offered by the same vendor, and should not be treated as having the same evidentiary reliability as plagiarism detection.