Fair and Sustainable Adoption of Artificial Intelligence in Educational Resource Management: Ethical Pathways for Resource-Constrained Contexts

Abstract

Educational planning, resource allocation, and institutional decision-making can be said to have in recent times been markedly scaffolded by the use of Artificial Intelligence (AI). A synthesis of neoteric literature grounded in the ethical dimensions of AI in educational resource management birthed the development of a conceptual framework for responsible AI implementation. Through thematic synthesis, and with supervision anchored on the PRISMA 2020 framework, 74 peer-reviewed studies and policy documents published between the span of 2021 and 2026 were retrieved from Scopus, Web of Science, ERIC, and Google Scholar and exhaustively analyzed. The findings provide arguably strong indication that responsible adoption of AI is fairly dependent on the interaction of the above-mentioned respective dimensions instead of any ethical principle in solitude. Algorithmic bias is arguably a consistent threat to fairness, which, in combination with transparency and accountability, emerges as mutually reinforcing in symbiotic likeness. Epistemic justice can be said to have close association with practitioner motivation and institutional readiness in blinding contrast to regional adaptation that appears more important for sustainable implementation than uncritical adherence to international frameworks. Although these relationships may be in need of empirical validation, the evidence provides a strong suggestion that contextual adaptation may better support long-term sustainability than universal frameworks applied without local interpretation. Integrating social cognitive theory, organizational systems theory, and epistemic justice scholarship, the study proposes a framework to guide ethical AI adoption in educational resource management, particularly within low-resource contexts such as Zambia.