Ethical and Governance Frameworks for Artificial Intelligence: A Systematic Literature Review

Abstract

Rapid proliferation of artificial intelligence (AI) in key domains such as healthcare, education, public services, and digital economies has heightened global commitment to developing strong ethical and governance standards. This systematic review, based on the PRISMA approach, synthesizes evidence from 22 peer-reviewed journal articles, white papers, and policy documents between 2020 and 2025. It examines governance frameworks, ethical concepts, regulatory approaches, sectoral application, and common challenges in operationalizing AI ethics internationally. The review sees wide confluence on core ethical values of fairness, transparency, accountability, explainability, and sustainability. There remains, however, wide variability in the application and institutionalization of these values across jurisdictions. The European Union’s AI Act provides a binding three-tiered risk system with centralized monitoring, whereas global institutions such as the World Health Organization (WHO) and the United Nations advocate high-level ethical frameworks but without statutory implementation. Regional efforts by ASEAN, Hong Kong SAR, and China reflect different states of maturity in the application of lifecycle audit and sector-level policy tools. Five recurring challenges are identified throughout literature: algorithmic bias, privacy risks of the data, patchwork of regulatory environments, corporate and state incumbents, and risks of labor market dislocation. Institutional models by the Alan Turing Institute and corporate players such as NTT DATA stress the value of building ethics from the outset of the life of AI. Educational and civil society voices add further calls for inclusive and collaborative governance frameworks. This paper concludes by advocating for the creation of an internationally coordinated AI regulation agency, cross-border certification schemes, ethics-by-design design methodologies, and adaptive governance mechanisms for the safe, equitable, and transparent application of AI technology.