Work

Welfare regimes and the adoption of artificial intelligence in healthcare

Year: 2026

Type: article

Abstract: This paper examines the integration of artificial intelligence (AI) into healthcare through the analytical lens of the political determinants of health (PDOH), arguing that AI is not a politically neutral technological innovation but a domain in which longstanding distributive logics of welfare regimes are reproduced and potentially intensified. As a conceptual, theory-building contribution, this paper develops a set of testable propositions about how welfare-regime logics may condition the adoption, regulation, and distributive effects of AI in healthcare. While existing scholarship on AI in healthcare has largely centred on ethical risks, algorithmic bias, privacy, and data governance, these discussions often overlook how political systems shape both the speed and equity of AI adoption. Drawing on Navaro et al.’s typology of political regimes —social democratic, Christian democratic, liberal, and ex-dictatorship—the paper argues that welfare state traditions and power resource configurations may structure how AI is developed, regulated, and what are its distributive consequences. Social democratic regimes, with strong universalist commitments, are likely to adopt AI cautiously but equitably; Christian democratic systems may integrate AI through corporatist negotiation and regulated pluralism; liberal regimes may accelerate market-driven AI innovation at the cost of uneven access; and ex-dictatorship states may face the dual pressures of modernization and institutional fragility, leading to uneven but politically charged trajectories of adoption. The paper concludes by arguing that regulatory progress on AI must be conceptualized in light of the political frameworks it comes to operate within and therefore, for AI in healthcare to be inclusive it may require a reaffirmation of the values of universal access.

Source: Social Science & Medicine

DOI: 10.1016/j.socscimed.2026.119938

Authors: Ritumbra Manuvie, Jeanne Pia Mifsud Bonnici, Mirjam Plantinga

Institutions: University of Groningen, University Medical Center Groningen

Language: English

Topic: Artificial Intelligence in Healthcare and Education

Subfield: Health Informatics

Field: Medicine

Domain: Health Sciences

SDGs: Reduced inequalities, Peace, justice, and strong institutions, Good health and well-being

Open Access status: hybrid

Estimated APC paid: $3,800