Hillary Segeren

This paper documents the methodological origin and development of the Segeren 2026 research programme in interaction-level AI governance. It is written as a methods and positionality statement — an account of how the research was actually generated, what questions drove it, and why the path taken produced the framework it did. The programme did not begin with artificial intelligence. It began with a question about how human beings make meaning — how symbolic systems have allowed people across history and culture to navigate complexity without collapsing it into false certainty. That question, followed far enough, produced a governance framework for AI interaction: named failure conditions, a preventive control architecture, and a forensic audit protocol applicable to any interaction record without access to the underlying model. None of it was designed in advance. Each element emerged from the previous one — from symbolic cognition through a live incident that had no name, through the slow work of naming what had happened, to external validation from within the academic lineage the research independently arrived at, and to a convergence finding that confirmed the problem is real enough to be found from two entirely different directions at once.