What Is the Impact of ChatGPT on Education? A Rapid Review of the Literature
Level 5 - mechanism / opinion, no new human data
Non-clinical design analogy: rapid review and qualitative content analysis of early exploratory literature without systematic pooling or controlled trials.
OpenAlex W4366420437 · doi:10.3390/educsci13040410
What was done
Authors conducted a rapid review of academic literature published during the first three months following ChatGPT's release (December 2022 to February 2023). Searches across academic databases and Google Scholar identified 50 articles, which were synthesized using qualitative content analysis (open, axial, and selective coding) to evaluate educational capabilities, use cases, and emerging concerns.
What was found
The abstract reports no quantitative performance metrics or pooled statistics. Qualitatively, ChatGPT's capabilities varied by discipline (categorized as outstanding in economics, satisfactory in programming, and unsatisfactory in mathematics). Identified applications included assisting instructors with course material generation and acting as a virtual tutor for students. Identified challenges included the generation of incorrect or fabricated information and the circumvention of plagiarism detection tools.
Why it matters
This review captures the initial wave of academic literature on generative AI in education, outlining early disciplinary strengths, use cases, and institutional risks identified immediately following ChatGPT's public release.
Limits
The review is limited to a narrow three-month timeframe (December 2022 to February 2023), reflecting very early, exploratory, and rapidly evolving literature. Methodological quality, sample sizes, and study designs of the 50 included articles are not reported in the abstract, and no quantitative synthesis was conducted.
Cited by
- supports ChatGPT was released in November 2022.