AI is now embedded in the systems that make decisions about people’s lives. It influences who gets hired, who is denied a loan, how diseases are identified, and who qualifies for social support. In response to its growing influence, the field of AI Ethics has emerged with a clear promise: to ensure that AI systems are fair, accountable, transparent, and aligned with human rights. It presents itself as a safeguard, a moral compass for intelligent technologies.
But as AI Ethics expands globally, a quiet question lingers beneath the surface: if this field claims to serve humanity, whose humanity is it actually designed for?
AI Ethics is not just a technical discipline. It is a global knowledge project. It shapes standards, informs governance frameworks, and influences national and international policies. What is defined as “ethical AI” does not remain inside academic journals, it travels across borders, institutions, and regulatory systems. In this sense, AI Ethics does not operate outside politics. It participates in what can be described as a techno-political order, where knowledge and power are produced together. The way ethical standards are defined shapes how authority is distributed, whose concerns are prioritized, and which values become global norms. Ethics, therefore, is not neutral. It is formed within existing structures of influence.
And those structures are uneven.
A recent study by the University of Cambridge analyzed a comprehensive database of 5,755 publications in AI Ethics from 1960 to June 2024. The study shows that a significant share of leading publications, conferences, and research institutions in AI Ethics are concentrated in Europe and North America. Scholars from these regions dominate citation networks, editorial boards, and funding structures. Meanwhile, institutions from Latin America, sub-Saharan Africa, and the Middle East and North Africa (MENA) remain largely absent from the core spaces where ethical standards are debated and defined.
This imbalance is not just about geography. It is about authority. When certain institutions consistently publish, cite one another, and shape the research agenda, they become recognized as the authoritative voices of the field. Their frameworks begin to appear universal. Their values begin to look neutral.
Much of the current literature relies on data sources, regulatory frameworks, and social concerns emerging from Western contexts. As a result, AI Ethics discussions often prioritize issues that are more visible within these environments. However, socio-economic realities differ significantly across regions. Ethical risks are not experienced uniformly, and concerns such as community-based decision-making models, local interpretations of privacy, and indigenous knowledge systems may receive less attention in mainstream debates. Recognizing these contextual differences is essential if AI Ethics is to remain globally relevant and practically effective.
This imbalance can be understood through the concept of epistemic injustice, a situation in which certain voices, perspectives, or knowledge systems are unintentionally marginalized within processes of knowledge production. In the context of AI Ethics, this may appear through limited recognition of expertise from underrepresented regions, unequal access to research platforms and funding networks, and narrower definitions of what counts as relevant ethical knowledge. Over time, such patterns influence whose concerns shape global standards and whose perspectives remain less visible.
These patterns have consequences beyond academia. When global AI governance frameworks are primarily informed by Northern scholarship, they may establish standards that are difficult to implement in Southern contexts. Technologies designed without local realities in mind can exacerbate existing inequalities. Policies imported without adaptation may generate resistance, mistrust, or regulatory gaps.
A constructive way forward for AI Ethics is exemplified by recent global efforts to build more inclusive knowledge platforms. One such initiative is the Independent International Scientific Panel on Artificial Intelligence, established by the United Nations in 2025 as the first global scientific body focused on AI. Comprising experts from diverse regions and disciplines, the Panel’s mandate is to produce independent, evidence-based assessments of AI’s risks, opportunities, and impacts for all nations. By making rigorous scientific insights accessible to a wide range of stakeholders and linking them to international dialogues on AI governance, this mechanism can help bridge knowledge gaps and support more equitable participation in shaping the ethical frameworks that govern this transformative technology.
So, when we ask the question of AI Ethics, it is not only about principles or technical safeguards. It is also about who gets to define the problem in the first place, whose concerns are prioritized, and whose realities shape the solutions.