Ethics as Infrastructure: Embedded Constraint in AI-Era Economic Systems

This paper argues that ethics in AI-era economic systems should be understood as embedded constraint architecture: a structural layer necessary for preserving legitimacy, feedback integrity, inclusion, and long-term system stability. As artificial intelligence becomes an optimization infrastructure capable of accelerating decisions, compressing labor needs, reorganizing production, and concentrating productive advantage, efficiency alone becomes an inadequate measure of progress. The paper develops a conceptual systems model linking optimization pressure, constraint strength, externalized harm, legitimacy, feedback integrity, and system stability. Drawing on systems theory, institutional economics, cybernetics, political legitimacy theory, and historical analogs from industrial labor, finance, public health, environmental governance, and safety engineering, the paper argues that high-impact infrastructures become durable only when their externalities are constrained and their risks are internalized into institutional design. The central contribution is a framework for treating ethics as infrastructure in AI-mediated economies. It proposes boundary conditions for AI-era economic governance, including minimum inclusion thresholds, harm internalization, participation density, concentration tolerance, transition absorbability, feedback trustworthiness, and time-horizon compatibility.