Systems Governed by Metrics Cannot Optimize Value

This volume consolidates a series of technical, strategic, and theoretical artifacts analyzing a shared structural phenomenon across contemporary algorithmic, institutional, and socio-technical systems: the substitution of human value optimization by metric-constrained, risk-averse governance. Rather than presenting isolated papers, this collection formalizes a coherent framework — Metric-Constrained Intelligence (MCI) — and documents its manifestations across recommender systems, creator economies, algorithmic governance, public policy, scientific evaluation, and AI oversight. Each artifact contained in this report is self-contained yet structurally coupled, allowing independent circulation while preserving a unified theoretical backbone. The objective is not speculative forecasting, but the formal documentation of failure modes, incentive misalignments, and governance risks already observable in deployed systems. This corpus is intended to function simultaneously as: ▫️a foundational reference framework, ▫️a prior-art record, ▫️and a modular base for academic, policy, and industry applications. Individual sections may later be expanded, published, or cited independently, while this record preserves the integrity and precedence of the complete framework.