Standing Eligibility versus Runtime Authorization: A Definitional Distinction for AI Governance

Abstract

Standing Eligibility versus Runtime Authorization distinguishes two governance instruments that are frequently collapsed in AI governance discussions. A standing-status determination takes a certified referent as its object, produces a status, and has continuing temporal effect. The referent may be an AI system, a defined configuration or deployment, an organization, or an AI management system. Where an applicable regime makes that status a condition of operation, market access, or participation, it functions as standing eligibility. A runtime action-release determination takes a specific proposed action as its object. The action is evaluated before execution under the applicable policy and authority state, and execution remains unavailable without an affirmative action-bound result. In the deterministic-governance-conformant form developed here, the determination produces an ALLOW, DENY, or ABSTAIN verdict together with an authorization artifact sufficient for independent reconstruction. The paper: formalizes the distinction between standing status and action-specific release by object, output, and temporal effect; documents the separation of these functions in aviation operations, regulated hospital medication use, and payment-card authorization, without treating their enforcement mechanisms as equivalent; distinguishes deterrent consequence, direct status enforcement, and action release; states a functional test for determining when scoped credentials, capability tokens, and request-level access controls cross into runtime authorization; separates runtime-authorization category membership from deterministic-governance conformance; distinguishes withholding standing status from an ABSTAIN verdict over a specific proposed act; and shows how certification, runtime authorization, and observability compose as three non-substitutable governance functions. The central claim is narrow: enforcement of standing eligibility does not, without an additional runtime dependency, produce an independently reconstructable pre-execution verdict over a specific proposed act. Certification establishes standing eligibility. Runtime authorization determines whether a specific proposed act may cross the execution boundary. Disclosure: FERZ, Inc. builds deterministic governance infrastructure for AI systems. This paper presents a category analysis and does not describe a FERZ product. Version 1.0, August 2026.