Axiomatic Identity Collapse Taxonomy formalizes collapse not as behavioural failure but as structural violation of identity conditions in artificial systems. Derived directly from the Four Axioms of Identity, this taxonomy provides an operational framework for detecting, classifying, and governing collapse across provenance, continuity, identity‑preservation, and Φ‑coherence layers. Each collapse mode corresponds to a specific invariant failure—ranging from missing or ambiguous provenance, unlawful or discontinuous transitions, unbounded deviation between model and organization, to structural, functional, or causal incoherence.The taxonomy extends naturally to multi‑agent and distributed systems, introducing meta‑collapse modes such as local and emergent collapse. It also outlines pre‑collapse regimes, threshold metrics, and early‑warning diagnostics, enabling engineers to build runtime monitoring, governance triggers, and safe self‑modification envelopes.Designed for engineering use, this work provides a rigorous, axiomatically grounded vocabulary for identity‑risk scoring, system‑of‑systems health assessment, and the construction of resilient AI architectures. It serves as a foundational reference for provenance engineering, AI safety, and high‑assurance autonomous systems.
Aure Ecker-Fils (2026) studied this question.