This systematic literature review examines how higher education institutions govern and implement generative artificial intelligence and what outcomes are reported for institutional efficiency and student learning. The review adheres to the PRISMA 2020 protocol. Searches across seven academic databases i.e., Scopus, Web of Science, Google Scholar, ERIC (Education Resources Information Center), IEEE Xplore, ScienceDirect, and ACM Digital Library covering 2020 to 2026 identified 239 records, with 18 duplicates removed. A total of 221 records were screened, yielding 101 exclusions. Subsequently, 120 full text reports were assessed, of which 70 were excluded with documented reasons. The final qualitative synthesis included 50 studies. Thematic findings indicate that effective adoption depends on governance maturity, characterized by coherent multi-unit oversight, clearly defined academic integrity and disclosure standards, and risk based data governance frameworks. The evidence demonstrates operational gains through automation and analytics, alongside improved student engagement and learning when implementation is supported by appropriate pedagogical scaffolding. However, persistent challenges remain, including risks of academic misconduct, inequitable access, privacy and bias concerns, and pronounced regional disparities, particularly in Global South contexts. Accordingly, institutions should prioritize embedded governance structures, equity safeguards, sustained faculty development, and continuous monitoring to ensure responsible, legitimate, and educationally meaningful deployment.
Bacalso et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: