Confronting AI: An Iatrogenic Problem for Democratic Governance

This paper critically examines the iatrogenic nature of regulatory efforts to govern artificial intelligence (AI) within democratic systems, demonstrating how attempts to mitigate the risks of AI often exacerbate the challenges they aim to address. AI’s integration into decision-making processes has brought ideals such as transparency, accountability, and inclusivity to the forefront as mechanisms for democratic oversight. However, these regulatory frameworks, while ostensibly designed to uphold democratic principles, paradoxically deepen reliance on algorithmic systems and amplify their authority over human judgment. Drawing on the concept of iatrogenesis—where interventions intended to cure instead cause harm—the paper reveals how democratic regulatory mechanisms entrench AI’s role in governance. Transparency initiatives, for instance, produce overwhelming amounts of data that require AI systems to analyze and interpret, shifting power further from human decision-makers to algorithmic processes. Similarly, accountability frameworks, which aim to maintain democratic oversight, impose procedural burdens that rely on AI’s efficiency and scale to manage complexity. These efforts, while seemingly democratic, reinforce AI’s centrality in governance, transforming oversight mechanisms into tools that perpetuate the very dynamics they seek to contain. The paper situates this critique within the broader trajectory of democratic governance, where ideals such as equality, freedom, and deliberation increasingly interact with algorithmic systems designed to optimize decision-making. It argues that these values, as traditionally conceived, often fail to address the structural challenges posed by AI’s bureaucratic logic. Instead of serving as a counterbalance, democratic values like transparency and accountability are operationalized in ways that align with the impersonal authority AI embodies. This paradox exemplifies what the paper identifies as the recursive trap of AI governance: regulating AI requires expanding its capabilities, which in turn magnifies its power and authority. The paper further explores the implications of this dynamic for democratic theory, addressing questions about the practicable limits of democracy in an age of technological complexity. Can democratic governance effectively constrain AI without reinforcing its dependence on algorithmic systems? Or does AI’s integration into governance signal a fundamental shift in the relationship between technology and democracy? By framing AI as an iatrogenic problem, the paper challenges dominant narratives that regulation alone can resolve AI’s democratic dilemmas, calling for a deeper interrogation of the values and assumptions underpinning efforts to govern technological systems. In the context of digital authoritarianism, the paper also highlights how AI-driven governance systems mirror dynamics often associated with autocratic regimes, including centralized control, surveillance, and depersonalized authority. Yet, these tendencies are not confined to non-democratic states; they reveal structural vulnerabilities within democratic systems themselves. This perspective urges scholars to reconsider the normative and conceptual frameworks used to analyze the intersection of democracy and technology. By engaging with the normative implications of AI for democratic values, this paper offers a critical intervention into contemporary debates on governance and technology. It contributes to understanding the recursive traps of AI regulation and invites reflection on the broader role of technology in reshaping democratic principles and practices in the digital age.