The Institutional Scaling Law: Non-Monotonic Fitness, Capability-Trust Divergence, and Symbiogenetic Scaling in Generative AI

ArXi:2603.14126v1 Announce Type: new Classical scaling laws model AI performance as monotonically improving with model size. We challenge this assumption by deriving the Institutional Scaling Law, showing that institutional fitness -- jointly measuring capability, trust, affordability, and sovereignty -- is non-monotonic in model scale, with an environment-dependent optimum N*(epsilon). Our framework extends the Sustainability Index of Han from hardware-level to ecosystem-level analysis, proving that capability and trust formally diverge beyond critical scale (Capability-Trust Divergence.