The Miulus Law: Epistemic Fitness as a Universal Constraint on Complex Self-Referential Systems

This paper introduces the Miulus Law, a universal information-theoretic constraint describing how the stability of any self-referential system depends on its ability to preserve verifiable signal against informational noise. The concept of epistemic fitness—the effective ratio of validated information to total noise multiplied by feedback reach—unifies failure modes across biology, cognition, society, and artificial intelligence. Empirical analysis of global governance and crisis data (1996–2024) reveals a consistent threshold below which systemic hazard accelerates non-linearly and collapse becomes inevitable. The study proposes that epistemic entropy, not material scarcity, is the primary driver of complex-system collapse and that continual verification work is the necessary condition for sustainable intelligence.