The Authorization Threshold: Defining the Standard for Governance in AI Systems

This paper defines the authorization threshold for governance in AI systems. It argues that governance is not a property of systems in general, but of specific decisions made under specific constraints. As AI systems increasingly initiate actions autonomously, operate across institutional and regulatory domains, and produce irreversible outcomes, governance must be evaluated at the level of action validity rather than system behavior. The paper distinguishes procedural approaches to AI governance (monitoring, alignment, filtering, logging, and observability) from the definitional conditions required for an action to qualify as governed. It proposes that an action is governed only if its authorization can be established prior to execution under explicit, evaluable constraints. Establishment is doing critical work in that definition: authorization is not checked but derived from defined constraints, bound to a specific action representation, and available for independent verification. To formalize this standard, the paper introduces the Canonical Action Frame, defines the distinction between interpretive systems and authorization systems, and draws a sharp line between evidence and proof. It argues that most governance architectures omit the critical transformation layer between policy and evaluation: the constraint representation layer, where human-readable rules are formalized into canonical, machine-evaluable constraints. The paper further defines seven conditions of governance validity and five dimensions of governance completeness: constraint completeness, semantic precision, policy binding, temporal binding, and independent replayability. It concludes by specifying the structure of a valid authorization artifact and proposing a testable standard for determining whether a decision was genuinely governed. This work is intended for researchers, regulators, enterprise buyers, compliance leaders, technical architects, and others evaluating what it means for AI governance to be formal, enforceable, and provable rather than merely observable or descriptive. Part of the FERZ deterministic governance research corpus. Within that corpus, this paper supplies the definitional layer: the threshold that separates governed from ungoverned action, the Canonical Action Frame, the evidence-versus-proof distinction, and the validity and completeness conditions any conforming system must satisfy. The proof requirements defined here connect directly to the Five Tests Standard (5TS), the vendor-neutral published standard specifying the Stop, Ownership, Replay, Escalation, and Provenance tests, with the Proof-Carrying Decision as its conformance proof object. Established input origin (the Provenance test) is treated as a normative requirement of proof-grade authorization: a proof is only as sound as the origin of the inputs it binds. Companion work in the corpus classifies governance approaches and their failure modes, establishes the five architectural dimensions of deterministic governance, and formalizes the semantic and type-theoretic substrates on which constraint representation depends.