Stable Authority Boundary (SAB) v1.0 — Conformance Specification

Stable Authority Boundary (SAB) v1.0 — Conformance Specification SAB v1.0 is a normative conformance specification that formalizes the requirements implied by the earlier work: Authority Contraction and Refusal as Safety Invariants in Autonomous Systems., and further defined in Authority, Refusal, and Resilience in Autonomous Systems: The Stable Authority Boundary as a Design Invariant It defines the minimum testable conditions for a Stable Authority Boundary: evidence MAY inform decisions but MUST NOT exercise authority to trigger execution. SAB prevents illegitimate execution by requiring authority proof at the boundary, prohibiting progress-by-default, and treating Refusal/Hold as correct behavior when authority is missing, degraded, ambiguous, revoked, or unverifiable. Designed to be implementable, testable, and auditable. SAB v1.0 specifies: A conformance outcome model: LEGITIMATE / AMBIGUOUS / REFUSAL REQUIRED / ILLEGITIMATE A minimum conformance test suite (SAB-REQ-01…SAB-REQ-10) with illegitimacy triggers (fail-open paths, boundary bypass, self-override, confidence-as-authority, and progress-by-default) Mandatory compliance artifacts (Authority Statement, Boundary Map, Refusal Policy, Override Policy, Traceability Ledger) Normative definitions for authority, evidence, boundary, execution, refusal/hold, and progress-by-default A required Compliance Statement template enabling third-party audit and reconstruction of the chain: inputs → evidence → boundary decision → authority proof → action/refusal Scope: SAB applies to any system capable of directly or indirectly triggering operational effect—software, hardware, workflow automation, decision-support systems, and AI-assisted tools—where evaluation outputs can be mistaken for authority, especially under degraded coordination, uncertainty, or partial failure. Canonical standard notice: This record is the sole canonical publication of SAB v1.0 (CSI: SAB-STD-v1.0-DF-2026). Redistribution and derivatives are permitted under CC BY 4.0; derivatives must cite the canonical DOI and must not present themselves as the canonical standard. Provenance: SAB v1.0 is derived from the author’s prior work, including Authority Contraction and Refusal as Safety Invariants in Autonomous Systems (Zenodo DOI: 10.5281/zenodo.18263456) and Authority, Refusal, and Resilience in Autonomous Systems: The Stable Authority Boundary as a Design Invariant (Zenodo DOI: 10.5281/zenodo.18431599) Keywords: Stable Authority Boundary (SAB), authority contraction, refusal as safety invariant, autonomous systems, distributed systems, failure modes, silent failure, system boundaries, authority boundaries, autonomous systems governance, authority limitation, safety invariants, refusal as safety behavior, system legitimacy, governance-first autonomy, execution authority, system authority, AI systems safety, distributed autonomous systems, resilience engineering, fail-safe design, refusal as safety behavior, system boundaries, authority vs control, governance-first autonomy, trustworthy autonomy, autonomous decision-making, failure modes, #EngineeringSeams,