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

TL;DR

This paper introduces the Institutional Scaling Law, revealing that AI performance related to institutional fitness is non-monotonic with model size and that orchestrated domain-specific systems can outperform generalist models in specific environments.

Contribution

It extends classical scaling laws by incorporating institutional factors, introduces the concept of capability-trust divergence, and proposes symbiogenetic scaling for optimized AI ecosystems.

Findings

Institutional fitness peaks at an environment-dependent optimal scale.

Capability and trust diverge beyond a critical model size.

Orchestrated domain-specific models outperform generalists in their niches.

Abstract

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 et al. (2025) from hardware-level to ecosystem-level analysis, proving that capability and trust formally diverge beyond critical scale (Capability-Trust Divergence). We further derive a Symbiogenetic Scaling correction demonstrating that orchestrated systems of domain-specific models can outperform frontier generalists in their native deployment environments. These results are contextualized within a formal evolutionary taxonomy of generative AI spanning five eras (1943-present), with analysis of frontier lab dynamics, sovereign AI emergence, and post-training alignment evolution from RLHF through GRPO. The Institutional Scaling Law predicts that the next phase transition will be driven not by larger models but by better-orchestrated systems of domain-specific models adapted to specific institutional niches.