AbstractThe MPHD (Propedeutic-Hermeneutic Method of Displacement) is a research program dedicated to the structural analysis of symbolic disorganization, decision-making collapse, and governance crises in human and computational systems. The framework proposes a meta-ontological architecture for mapping affective displacement, auditing symbolic coherence, and restoring decision axes in complex environments. In the context of Artificial Intelligence, the MPHD operates as a Meta-Audit layer, offering ontological supervision for mitigating hallucinations and supporting safety alignment in Large Language Models (LLMs). Rather than focusing on code-level optimization, the system introduces a cartography of displacement capable of identifying structural inconsistencies in synthetic responses and decision flows. Organized into modular components (MPHD-0 to MPHD-III), the program integrates ontology, phenomenology, and symbolic mediation to support applications in AI governance, institutional crisis mediation, and complex clinical contexts. This Overview presents the conceptual architecture of the program, establishing a foundation for academic validation, technological implementations, and formal research partnerships.
Antônio Rodrigues de Miranda Júnior (Sun,) studied this question.