One year in the classroom with ChatGPT

Empirical insights and transformative impacts

Bibliographic Data

ID22166173
AuthorsFeng Guo (0000-0003-3584-154X, University of Tennessee at Chattanooga, corresponding author), Tian Li (0000-0002-8281-4459), Tianrui Li (0000-0001-7780-104X, University of Tennessee at Chattanooga), Christopher J L Cunningham (0000-0001-6889-9962, University of Tennessee at Chattanooga)
Year2025
Volume10
Publication date2025-05-27
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueFrontiers in Education (JOURNAL)
Journal identifiersISSN: 2504-284X • E-ISSN: 2504-284X
PublisherFrontiers Media SA (PUBLISHER • CH)
DOI10.3389/feduc.2025.1574477
OpenAlexW4410772574
LanguageEN
Citations received1
References cited153

Generative Artificial Intelligence (GAI), such as OpenAI’s ChatGPT, has rapidly emerged as a transformative tool in higher education, offering opportunities to enhance teaching and learning. This paper describes the design and implementation of ChatGPT-integrated curriculum activities, featuring coding learning in psychology and conceptual discussions in physics, and presents the findings of a year-long experimental study in both types of classrooms. Our findings suggest that students generally found ChatGPT easy to use and beneficial to their learning, reporting improved confidence, motivation, and engagement. However, its ability to address individual needs or replace instructors was viewed less favorably. Comparative analyses showed that coding activities in psychology led to higher levels of activity satisfaction and perceived usefulness of ChatGPT compared to the more abstract discussion activities in physics. While graduate students were more enthusiastic about using ChatGPT for skill acquisition than undergraduates, demographic factors such as gender, race, and first-generation college status showed no significant influence on such perceptions. Meanwhile, instructors’ reflections emphasize the importance of thoughtful integration, technical support, and pedagogical balance to maximize GAI’s potential while mitigating its limitations. Recommendations for integrating GAI into teaching practices and future research directions are discussed, contributing to the evolving discourse on GAI’s role in transforming modern classrooms

Data science · Empirical research · Mathematics education · Pedagogy · Statistics · Transformative learning · Artificial Intelligence in Healthcare and Education · Computer Science · COVID-19 diagnosis using AI · Machine Learning in Healthcare · Mathematics · Psychology

Tools

Open DOIOpen Access