Generative AI in Higher Education: A Systematic Literature Review

Abstract

Background: Generative Artificial Intelligence (GenAI) technologies such as ChatGPT, Bard, and Claude have emerged as transformative forces in higher education, reshaping teaching, learning, assessment, and research practices. While their integration offers significant potential for personalisation, pedagogical innovation, and productivity enhancement, it simultaneously raises complex concerns regarding ethics, academic integrity, data privacy, and institutional readiness. The fragmented nature of existing research, particularly across Global South contexts such as South Africa, highlights the need for a comprehensive synthesis to guide responsible and equitable GenAI adoption. Purpose: This study systematically reviews recent empirical and conceptual research to examine GenAI readiness, adoption determinants, ethical implications, and regional disparities within higher education. It further identifies theoretical limitations within established frameworks, Technology Acceptance Model (TAM), Technology-OrganisationEnvironment (TOE), Technology Readiness Index (TRI), and Theory of Planned Behaviour (TPB) and proposes a hybrid conceptual framework integrating socio-cultural and ethical dimensions. Methods: A PRISMA-guided systematic literature review was conducted across Scopus, Web of Science, ERIC, IEEE Xplore, and Google Scholar. Out of 192 identified studies, 18 met the inclusion criteria (2023-2025). Thematic synthesis combined inductive and deductive coding to map patterns across readiness, adoption predictors, ethical issues, and policy trends. Results: Findings reveal four key readiness dimensions: technical access, psychological preparedness, ethical literacy, and institutional support. Predictors of adoption included perceived usefulness, ease of use, selfefficacy, and leadership support, while barriers centred on ethical uncertainty, data privacy concerns, and infrastructural limitations. Conclusions: A hybrid TAM, TOE, TRI, TPB framework, extended with socio-cultural and ethical components, is proposed to explain GenAI adoption holistically. The study recommends targeted AI literacy, ethical governance, and context-sensitive policies to foster equitable and sustainable integration of GenAI in higher education.

Ähnliche Arbeiten