The bright highlights and the dark shadows of artificial intelligence in higher education

Abstract

Background Artificial Intelligence (AI) is reshaping higher education, prompting critical debate about its implications for teaching and learning. While AI offers substantial opportunities, its integration also raises concerns related to academic integrity, skill development, bias, privacy, and educational inequality. Methods This study employed a Systematic Conceptual Review (SCR) of 105 studies drawn from Google Scholar, ERIC, and Scopus and published between January 2021 and July 2025. The literature was analysed through thematic synthesis to examine the dual impact of AI in higher education and the roles of educators, students, and institutions in AI-mediated environments. Results The analysis identified a tension between “Bright Highlights,” including personalised learning, operational efficiency, and enhanced accessibility, and “Dark Shadows,” including risks to academic integrity, loss of basic skills, algorithmic bias, data privacy, and increased educational inequality. The synthesis further identified four factors that mediate these outcomes: pedagogical, ethical and integrity, technical and quality assurance, and sociostructural mediation. These findings informed the development of the Bright-Shadow Integration Model, which conceptualises AI outcomes as shaped by the interaction between AI capabilities and these mediating factors. Conclusions The findings indicate that the effects of AI in higher education are not determined by the technology itself but by how its capabilities are mediated within educational contexts. A proactive approach involving students, educators, institutions, and policymakers, supported by clear policies, pedagogical redesign, ethical guidelines, and attention to equity, is therefore essential for harnessing AI’s potential while mitigating its risks.