Taxonomy of Artificial Standing Failure

This paper presents a taxonomy of artificial standing failure. It classifies the distinct, exhaustive modes by which artificial epistemic agents lose standing, showing that failures are structural rather than behavioral or performance-based. Each class is defined by violations of admissibility, scope invariance, or standing conservation, and all apparent hybrids or partial failures are reduced to these categories by exhaustion. The taxonomy demonstrates why standing failure is binary, non-local, and non-repairable across architectures. No diagnostic algorithms, remediation steps, or implementation guidance are provided; the result is classificatory and necessity-based.