Sung Jun (Jun) Han

Recent surveys reveal widespread disillusionment with electoral representative democracy due to elite capture, prompting interest in lottocracy—randomly selecting legislators to enhance democratic authenticity and epistemic quality of decisions. However, current lottocratic models face two challenges: accountability, as randomly selected legislators lack the reelection pressure to act in the public interest, and authenticity, as citizens might blindly defer to a random few. This essay introduces the Madisonian lottocracy, a model that preserves lottocracy's benefits while overcoming these hurdles. In this model, multiple single-issue legislatures serve a filtering role rather than making final decisions. Randomly selected legislators deliberate on policy options, which are then presented to a jurisdictional referendum. This design instills a form of electoral accountability that incentivizes responsive action and allows citizens to critically assess refined options instead of deferring unconditionally.

This paper examines the integration of Artificial Intelligence (AI) into democratic governance, focusing on the tension between democracy’s epistemic shortcomings—often manifested as voter ignorance—and AI’s capacity to improve decision-making. Building on the Input-Process-Output (IPO) model, the paper distinguishes AI applications into four categories based on the democratic source of their inputs (i.e., whether they originate from the citizenry) and the binding nature of their outputs (i.e., whether AI decisions carry legal or authoritative weight). Each category—democratic binding AI, undemocratic binding AI, democratic unbinding AI, and undemocratic unbinding AI—is then evaluated against core democratic elements: inclusive and equal participation, quality of decisions, deliberation, and the autonomy of citizens to set the political agenda. While some undemocratic binding AI risks centralizing power into the hands of a few, certain forms of AI, such as AI advisers, AI delegates with deliberative consent, and AI nudger, can enhance democratic processes by helping citizens overcome epistemic barriers, refine their political views, and participate more effectively in governance. The paper concludes that carefully implemented AI has the potential to enhance democratic governance while preserving its core ideals.