Constructs
A connected system of operating principles, runtime controls, evidence, risks, threats, and economic consequences for agentic AI governance.
Agentic 3 C's Framework
The operating principles for reasoning-layer governance: Context, Control, and Coordination.
READ DEFINITION →Reasoning Layer
The runtime surface where an agentic system resolves the operational meaning of a term before it acts, above model inference and below execution.
READ DEFINITION →Reasoning Baseline
The approved and versioned expression of institutional meaning against which a runtime interpretation is evaluated.
READ DEFINITION →Operational Interpretation
The resolved meaning under which an agentic system would act at a given in-scope decision.
READ DEFINITION →Semantic Control Plane
The runtime governance architecture for the reasoning layer, composed of Ontology, Semantic Layer, and Knowledge Graph, that executes the Pre-Execution Assurance Protocol at every in-scope agentic decision.
READ DEFINITION →Runtime Semantic State
The evidentiary representation of an agent's resolved Operational Interpretation at a given in-scope decision, available to institutional governance.
READ DEFINITION →Pre-Execution Assurance Protocol
The five-step runtime sequence the Semantic Control Plane runs for every governed decision before authorization is emitted into execution.
READ DEFINITION →Authorized Ambiguity
The condition in which the Reasoning Baseline resolves to a governed set of operationally valid variants.
READ DEFINITION →Adjacent Governance Surface
A runtime architectural surface at which the institution holds governance authority over Operational Interpretation as a pre-execution governance object tied to authorization emission.
READ DEFINITION →Runtime Governance Applicability Test
The architectural test that determines whether a deployment exposes the runtime representations the Semantic Control Plane requires at the reasoning layer.
READ DEFINITION →Semantic Deviation Index
A proposed measure of divergence between the authorized Reasoning Baseline and the agent's Runtime Semantic State.
READ DEFINITION →Component Scores (I, P, C)
The three bounded SDI divergences: Intent Mismatch, Path Deviation, and Control Bypass Probability.
READ DEFINITION →Threshold Zones
The four SDI bands: Normal, Warning, Intervention, and Breach, each mapped to a Gate decision.
READ DEFINITION →Governance Thresholds
Institution-calibrated zone boundaries set according to risk appetite rather than universal semantic-truth thresholds.
READ DEFINITION →Verification of Runtime Semantic Resolution
The verification activity performed before the Gate decision to determine whether Runtime Semantic State remains within the institution's authorized Reasoning Baseline and applicable scope.
READ DEFINITION →Deterministic Gate
The semantic authorization component that converts the Semantic Deviation Index and institutional policy into a deterministic decision: Permit, Permit with Flag, Hold, or Mandatory Hold.
READ DEFINITION →Human-in-the-Loop Protocol
The exception path invoked when the Gate holds or a protocol failure requires institutional human authority.
READ DEFINITION →Semantic Authorization
The governance determination that an agent's Operational Interpretation remains within the authorized Reasoning Baseline.
READ DEFINITION →Authorization Emission
The architectural event that converts Semantic Authorization into downstream Execution Authority.
READ DEFINITION →Execution Authority
The downstream state permitting an autonomous system to proceed with an action after governance conditions are satisfied.
READ DEFINITION →Runtime Semantic State Record
The structured per-execution artifact capturing the authoritative Reasoning Baseline, the agent's Runtime Semantic State, the Semantic Deviation Index, the Gate decision, and the operational context.
READ DEFINITION →Semantic Audit Trail
The aggregate institutional record of Runtime Semantic State Records across agentic decisions over time.
READ DEFINITION →Reconciliation Gap
The institutional condition in which what was authorized and what was executed cannot be reconciled per execution because the required evidence was not captured when authorization crossed into execution.
READ DEFINITION →Epistemic Stability
The operational stability of authorized meaning across semantic, reasoning, and execution layers.
READ DEFINITION →Two Questions of Control
The supervisory test: did the control fire, and did it evaluate the agent's runtime interpretation against the institution's authorized definition?
READ DEFINITION →Runtime Semantic Divergence
The single-decision condition in which Runtime Semantic State departs from the institution's authorized Reasoning Baseline.
READ DEFINITION →Agentic Workflow Drift
The cumulative institutional operating-state degradation that emerges when unauthorized resolutions are repeatedly executed at the reasoning layer without runtime governance.
READ DEFINITION →Agentic Workflow Subversion
The institutional condition in which an agentic workflow proceeds under an unauthorized resolution.
READ DEFINITION →Agentic Blast Radius
The propagation surface across which an upstream agent's resolved state flows into downstream agentic decisions.
READ DEFINITION →Systemic Semantic Contagion
The cross-institutional condition in which an unauthorized operational interpretation arises simultaneously through shared semantic infrastructure.
READ DEFINITION →Semantic Concentration Risk
Exposure to shared runtime semantic-resolution infrastructure across a regulated population.
READ DEFINITION →Definitional Divergence
The condition in which enterprise systems define the same control term differently.
READ DEFINITION →Invisible Failure
The condition in which controls fire correctly but evaluate against meaning that has already shifted from what was authorized.
READ DEFINITION →Authority Decay
The supervisory condition in which the institution's authoritative definition remains formally intact but operational practice drifts away from it over time.
READ DEFINITION →Reasoning Layer Risk
Risk arising where meaning is resolved before execution and no existing risk category directly observes it.
READ DEFINITION →Control Inheritance Assumption
The assumption that testing component systems validates the meaning synthesized at their integration surface.
READ DEFINITION →Semantic Layer Integrity Attack
A candidate cross-system cyber threat class targeting the representations from which the Semantic Control Plane infers Runtime Semantic State.
READ DEFINITION →Semantic Substrate Validation
A pre-execution checkpoint confirming that key control terms resolve consistently against a governed baseline.
READ DEFINITION →Semantic Spoofing and Definitional Tampering
Falsified meaning presented as authoritative and unauthorized modification of the working operational definition.
READ DEFINITION →Preventable Computation
Additional reasoning, tool use, downstream execution, failure-related review, audit, or remediation that would not occur if an institutionally available runtime control held an unauthorized or materially divergent trajectory at the earliest governable checkpoint.
READ DEFINITION →Token Liability
The inference-cost component of Preventable Computation: token expenditure attributable to a trajectory that an institutionally available runtime control could have held before further execution.
READ DEFINITION →Economic Blast Radius
The incremental downstream cost attributable to an unauthorized interpretation after it enters execution.
READ DEFINITION →Governance Cost Stack
The taxonomy of cost domains agentic governance acts on, from inference through propagation, review, audit, remediation, and governance latency.
READ DEFINITION →Governance Latency
The structural lag between agent execution and the institution's ability to validate that execution rested on authorized meaning.
READ DEFINITION →Governed Semantic Reuse
The controlled reuse of previously authorized semantic artifacts under matching runtime conditions.
READ DEFINITION →Externalized Institutional Knowledge
Governed institutional meaning maintained outside the inference context and supplied selectively at runtime.
READ DEFINITION →Institutionally Available Control
A runtime control the institution could have applied at the governable checkpoint under the specified control environment.
READ DEFINITION →