JMIR Publications

ABSTRACT

Artificial intelligence (AI) is increasingly embedded in the institutions and environments that shape health, yet current frameworks for understanding its implications for health equity remain underdeveloped. The social determinants of health (SDoH) tradition provides a strong foundation, and recent work on digital determinants of health has begun to address the health implications of digital transformation. However, AI warrants distinct conceptual attention because it simultaneously functions as a sector-specific determinant, a cross-sector policy object, and an infrastructural technology that can reorganise how opportunities, resources, and risks are distributed across multiple domains of social life. This Viewpoint proposes a redistribution-translation-accumulation (RTA) framework for analysing how AI may contribute to the reproduction of health inequity. Redistribution captures how AI reshapes the distribution of health-relevant resources and opportunities–including education, employment, and income–while AI itself becomes an unequally distributed determinant. Translation describes how AI changes the pathways through which social positions are converted into health outcomes: proxy-based decision rules can formalise historical inequities, diagnostic algorithms may perform unevenly across populations due to training data underrepresentation, and AI-mediated information environments can alter institutional responsiveness and exposure to misinformation. Accumulation addresses how AI-driven feedback loops and institutional embedding can concentrate advantage and disadvantage over time, as small initial differences compound through repeated human-AI interactions–often without users' awareness that their judgements are being altered. Together, these three mechanisms capture AI's capacity to act not merely as a new health exposure, but as a force that can reconfigure the generative processes underlying health inequity. The framework has direct implications for governance: current approaches such as the EU AI Act's Fundamental Rights Impact Assessment and Canada's Algorithmic Impact Assessment address important dimensions of AI accountability, but do not systematically evaluate distributional health consequences. I propose a Distributional Impact Assessment, structured around the RTA dimensions, as a complementary tool for equity-oriented AI governance.