Mapping and Comparative Analysis of Ethical Frameworks in Artificial Intelligence: Convergences and Gaps

Abstract

The rapid expansion of Artificial Intelligence (AI) has raised ethical and regulatory concerns, highlighting the need for robust frameworks to ensure its responsible development. This research conducted a systematic mapping and a comparative analysis of the main international ethical and regulatory frameworks, including proposals from the OECD, the European Union (AI Act), IEEE, ISO/IEC, Canada, the United States, and China. The study is based on the hypothesis that, although there is convergence around fundamental principles such as transparency, fairness, and responsibility, significant gaps remain in implementation, regulatory coherence, and cultural adequacy. The objective of this study is to identify convergences and divergences among the frameworks, assess their implementation mechanisms, and highlight practical challenges. The methodology adopted is qualitative, based on documentary analysis of normative texts, reports, and specialized literature, following PRISMA guidelines for systematic reviews. The results reveal growing conceptual alignment among jurisdictions but also highlight fragmentation in the operationalization of ethical principles, enforcement capacity, and socio-technical responsiveness. Gaps such as the absence of binding mechanisms, technical opacity, and low public participation are particularly critical. It is concluded that the effectiveness of these frameworks depends on the articulation of legal norms, technical standards, and participatory processes, as well as the inclusion of diverse global perspectives. The study proposes guidelines for an adaptable, responsible, and culturally sensitive ethical governance model.