Ability of AI detection tools and humans to accurately identify different forms of AI-generated written content

Adam Cheng; Yiqun Lin; Gabriel Reedy; Christine Joseph; Samantha Wirkowski; Viviane Mallette; Vikhashni Nagesh; David Krieser; Aaron Calhoun

doi:10.1186/s41077-025-00396-6

Ability of AI detection tools and humans to accurately identify different forms of AI-generated written content

Adv Simul (Lond). 2025 Nov 22;10(1):66. doi: 10.1186/s41077-025-00396-6.

Authors

Adam Cheng¹, Yiqun Lin², Gabriel Reedy³, Christine Joseph⁴, Samantha Wirkowski⁵, Viviane Mallette⁶, Vikhashni Nagesh⁷, David Krieser⁸, Aaron Calhoun⁹

Affiliations

¹ Departments of Pediatrics and Emergency Medicine, Cumming School of Medicine, KidSIM Simulation Program, Alberta Children's Hospital, University of Calgary, 28 Oki Drive NW, Calgary, AB, T3B 6A8, Canada. chenger@me.com.
² KidSIM Simulation Program, Alberta Children's Hospital, Calgary, Canada.
³ Clinical Education, Faculty of Life Sciences and Medicine, King's College London, London, England.
⁴ Division of Pediatric Critical Care Medicine, Norton Children's Hospital, University of Louisville School of Medicine, Louisville, USA.
⁵ Fellow of Pediatric Critical Care Medicine, University of Louisville School of Medicine, Louisville, KY, USA.
⁶ Department of Pediatrics, Section of Neonatology, Centre Hospitalier Universitaire Sherbrooke, Sherbrooke, Canada.
⁷ Department of Pediatrics, Cumming School of Medicine, University of Calgary, Calgary, Canada.
⁸ Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, Parkville, Australia.
⁹ Pediatric Critical Care, University of Louisville School of Medicine, Louisville, KY, USA.

Abstract

Background: The increasing use of artificial intelligence (AI) by scholars presents a pressing challenge to healthcare publishing. While legitimate use can potentially accelerate scholarship, unethical approaches also exist, leading to factually inaccurate and biased text that may degrade scholarship. Numerous online AI detection tools exist that provide a percentage score of AI use. These can assist authors and editors in navigating this landscape. In this study, we compared the scores from three AI detection tools (ZeroGPT, PhraslyAI, and Grammarly AI Detector) across five plausible conditions of AI use and evaluated them against human assessments.

Methods: Thirty open access articles published in the journals Advances in Simulation and Simulation in Healthcare prior to 2022 were selected, and the article introductions were extracted. Five experimental conditions were examined, including: (1) 100% human written; (2) human written, light AI editing; (3) human written, heavy AI editing; (4) AI written text from human content; and (5) 100% AI written from article title. The resulting materials were assessed by three open-access AI detection tools and five blinded human raters. Results were summarized descriptively and compared using repeated measures analysis of variance (ANOVA), intraclass correlation coefficients (ICC), and Bland-Altman plots.

Results: The three AI detection tools were able to differentiate between the five test conditions (p < 0.001 for all), but varied significantly in absolute score, with ICC ranging from 0.57 to 0.95, raising concerns regarding overall reliability of these tools. Human scoring was far less consistent, with an overall accuracy of 19%, indistinguishable from chance.

Conclusion: While existing AI detection tools can meaningfully distinguish plausible AI use conditions, reliability across these tools is variable. Human scoring accuracy is uniformly low. Use of AI detection tools by scholars and journal editors may assist in determining potentially unethical use but they should not be relied upon alone at this time.

Keywords: Academic writing; Artificial intelligence; ChatGPT; Detection; Ethics; Large language models.