Recalibrating Responsible AI: Non-Western Ethics, Sustainable Machine Learning, and Human Oversight

Abstract

The proliferation of smart systems in all spheres of society has called attention to the ethical issues of responsibility, fairness, and transparency associated with the development and deployment of AI at a global level. The more that machine learning becomes enmeshed in processes of decision-making, the more new ethical dilemmas that reflect the possibility for algorithmic bias, information vulnerability, socio-cultural exclusion, and environmental considerations begin to surface. This review provides an organised, humane, systematic examination of responsible AI, connecting ethical considerations with pragmatic concerns as derived from existing models, policies or problems in execution. Apart from standard controversies, the contribution of the paper is to single out little explored concerns that are related to non-Western ethical perspectives, sustainable AI practices, and the ethical reasoning of superintelligent systems. It is a critique of the existing limitations of ethical codes, which put forward ideas and directions to connect innovation and social good. By bringing in alternative perspectives and interdisciplinary thinking, it presents a rich base for the ethical recalibration of intelligent systems to work for human betterment, in a more inclusive, equitable and forward-looking manner.

Keywords: Responsible artificial intelligence; Algorithmic fairness and non-discrimination; Transparency and accountability; Data privacy and governance; Global AI governance

DOI: 10.7176/JIEA/15-2-09

Publication date: September 28^th 2025