Triangulated AI Inquiry (TAI): A Methodology for Human-AI Collaborative Theoretical Development

We present a novel methodology for theoretical development that leverages multiple artificial intelligence systems in structured dialogue, guided by human insight and integration. This approach—which we term "Triangulated AI Inquiry" (TAI)—addresses limitations of both traditional solo theoretical work and simple human-AI interaction. By employing multiple AI systems in distinct roles (development, pressure-testing, and independent review), TAI produces theoretical frameworks that are more rigorously examined, more robustly defended, and more honestly circumscribed than either human or AI working alone could achieve. We document the methodology through two case studies: the development of Chaotic Gravitational Criticality (CGC), a framework for substrate-dependent consciousness, and Zentrafuge, an AI mental health companion platform for military veterans. We provide practical guidance for researchers seeking to employ similar methods. We argue that TAI functions not merely as a productivity tool, but as a new epistemic instrument: one that externalises conceptual disagreement and forces explicit adjudication of theoretical choice points. KEYWORDS: human-AI collaboration research methodology large language models epistemic instruments multi-agent systems theoretical development conceptual friction triangulated inquiry