Purpose This paper aims to examine how artificial intelligence (AI) is transforming the epistemic foundations of higher education. It conceptualizes AI not as a pedagogical tool but as an epistemic agent that fundamentally redefines how knowledge is created, validated and governed in academic institutions. Design/methodology/approach An integrative literature review was conducted across Scopus, Web of Science and Education Resources Information Center (ERIC) databases, covering interdisciplinary scholarship from 2018 to 2025. Thematic synthesis was employed to analyze sources spanning education, philosophy, critical algorithm studies and digital ethics. Findings The review identifies three interconnected transformations: AI assumes epistemic co-agency in knowledge production, algorithmic governance redistributes institutional power toward automated systems and current ethical frameworks remain too abstract for classroom-level implementation. These shifts signal a move from cognitive offloading to cognitive reconfiguration, requiring fundamental reconceptualization of higher education. Research limitations/implications As a conceptual synthesis, empirical validation of the proposed “algorithmic university” framework is needed. Future research should examine how epistemic authority is negotiated in AI-mediated classrooms and how algorithmic bias manifests in educational contexts. Practical implications Universities should establish layered AI governance structures, invest in faculty development for epistemic literacy and prioritize pedagogically driven AI procurement over administrative optimization. Social implications Without intentional design prioritizing equity and transparency, AI integration risks exacerbating educational inequalities and concentrating institutional power in opaque commercial systems. Originality/value This paper advances the concept of the “algorithmic university” as a framework for understanding AI's systemic impact, moving discourse beyond tool-based integration toward epistemic and institutional transformation.
Babu George (2026) studied this question.