This work formalizes Individual-Scale AI Orchestration (I.S.A.O.) as a reproducible methodological framework for coordinating human decision authority across heterogeneous, non-deterministic AI systems. Building on the Interactive Intelligence Systems (IIS) framework, intelligence is treated as an emergent property of structured human–AI interaction rather than a function of model capability alone. Within this paradigm, Adaptive Intelligence Systems (AIS) describe systems whose effective performance depends on feedback, verification, and distributed decision authority across interaction loops. Over an approximately nineteen-month development period (June 2024–January 2026), the methodology was validated through multiple externally verifiable outcomes, including professional certification attainment, complex financial remediation, and the recursive completion of this manuscript using I.S.A.O. itself. The core contribution is a vendor-agnostic orchestration protocol integrating multiple large-language-model platforms into a fault-tolerant Distributed Intelligence Mesh System (D.I.M.S.), demonstrated to sustain operational continuity during real-world platform disruptions. This version (v1.5) incorporates expanded framework formalization, cross-paper integration with the IIS framework, and refined documentation of methodological scope, evaluation boundaries, and human-in-the-loop accountability mechanisms. The work demonstrates that methodology, not computational scale, is the primary constraint in independent, publication-grade AI research.
E. Martin Browne (2025) studied this question.