Practical evidence review explains AI-writing and similarity scores in academic settings, suggesting interpretation methods.
This practical evidence review explains why Turnitin’s AI-writing indicator and similarity score must be interpreted separately. The AI-writing indicator estimates the share of qualifying long-form prose that may have been generated or altered by AI, while the similarity score measures text overlap with sources in selected databases. Neither percentage, on its own, establishes academic misconduct. The report summarizes current Turnitin guidance, explains the below-20% asterisk, maps four common score combinations, and provides a defensible review workflow for students and educators. It also discusses limitations identified in peer-reviewed research on AI-text detection. Prepared by PaperCheck, an independent pre-submission review platform: https://aiturnitinchecker.com/ PaperCheck is not affiliated with, endorsed by, or sponsored by Turnitin LLC. Turnitin is a registered trademark of Turnitin LLC.
No takes yet. Share an insight, caveat, or question.
PaperCheck (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: