Ethics, AI, and Vulnerable Populations: A Scientometric Analysis of Global Regulatory and Governance Frameworks

Abstract

The increasing integration of artificial intelligence (AI) into socio-economic structures raises major ethical challenges, particularly regarding the risks of exacerbating inequalities and systemic discrimination. This study provides a comprehensive global scientometric analysis of scientific literature devoted to AI regulation and governance frameworks concerning vulnerable populations from 2000 to 2026. A rigorous corpus of 2,563 peer-reviewed articles extracted from the Scopus database (restricted to Social Sciences, Economics, and Management) was computationally mapped using science-mapping techniques. Keyword co-occurrence and co-citation analyses highlight a critical conceptual transition: global research is evolving from an initial focus on technical data compliance toward imperatives of algorithmic justice, institutional accountability, and social inclusion. The study identifies key intellectual structures and author networks shaping public policies for the protection of marginalized groups. By synthesizing these research trajectories, this paper offers a strategic theoretical framework to design ethical AI governance capable of reconciling technological innovation, social equity, and distributive justice.