2025/05/27 by David Tyler, Suzanne St. George, Analisa Gagnon +1 · 1 voice
Computer Science · Social Sciences · #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #Online Learning and Analytics
paper · doi:10.1080/10511253.2025.2506435
openalex publication_date 2025/05/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/15
This study explores whether criminal justice educators can differentiate between AI- and student-generated content in essay responses and investigates their rationale for categorizing authorship. A mixed-methods experimental vignette design was used, with 54 criminal justice educators assessing nine short-answer essay responses. Regression and qualitative content analysis are employed to examine educators’ accuracy and reasoning. Educators accurately identified student-generated content 81% of the time, but only accurately identified AI-generated content 59% of the time. Educators’ reasoning primarily focused on writing mechanics (e.g. grammar, style) to guide authorship assessments. Criminal justice educators are adept at identifying student work but less accurate with AI-generated content. Findings suggest educators’ heuristics could be improved by focusing more on the content and creativity rather than mechanics.