Evaluator as Self-Detector of AI-Generated Assignments in Nursing

Introduction: The recent rise in Artificial Intelligence (AI), and specifically the release of Chat Generative Pre-Trained Transformer (ChatGBT) has garnered significant attention among academic circles. ChatGBT is an AI-powered chatbot trained to provide human-like responses to text-based queries. Despite its growing popularity, there is a lack of research on whether educators can distinguish between AI-generated and student-crafted assignments.
Goals and Methods: The goal of this study is to determine whether there are noticeable differences between AI-generated and student-crafted assignments. Four students were selected from a second-year undergraduate nursing theory course in Toronto, Canada. The students submitted two copies of an assignment: one written by them, and one that was AI-generated. The assignments were anonymized by the research assistant and graded by the principal investigator (PI). The PI used a reflective methodology to describe their grading experience.
Results: There were noticeable differences between the AI-generated and student-crafted assignment particularly in the areas of writing style, paragraph structure, references used and the referencing style accuracy. This also led to differences in grades students received on both assignment versions.
Conclusions: The findings from this study provide strategies on how evaluators can become self-detectors of AI-generated assignments versus student-crafted assignments. This study provides suggestions on adjusting assignment guidelines to encourage students to cultivate critical thinking and analysis independent of AI. Ultimately, this study can serve as an important resource for educators navigating the use of AI in nursing education and implications of such use on academic integrity.