Governance artifacts
Governance files brought into scope by this page
This page is anchored to published surfaces that declare identity, precedence, limits, and the corpus reading conditions. Their order below gives the recommended reading sequence.
EAC registry
/.well-known/eac-registry.json
Normative registry for admissibility of external authorities in the open web.
- Governs
- Admissible relations, receivable authorities, and conflict arbitration.
- Bounds
- Abusive merges, copied authority, and unqualified silent arbitration.
Does not guarantee: Describing a graph or registry does not make an exogenous source endogenous truth.
Admissible exogenous claims
/eac-claims.json
Surface that bounds receivable families of exogenous claims.
- Governs
- Admissible relations, receivable authorities, and conflict arbitration.
- Bounds
- Abusive merges, copied authority, and unqualified silent arbitration.
Does not guarantee: Describing a graph or registry does not make an exogenous source endogenous truth.
EAC conflicts
/eac-conflicts.json
Surface for exogenous conflict arbitration and its resolution conditions.
- Governs
- Admissible relations, receivable authorities, and conflict arbitration.
- Bounds
- Abusive merges, copied authority, and unqualified silent arbitration.
Does not guarantee: Describing a graph or registry does not make an exogenous source endogenous truth.
Complementary artifacts (3)
These surfaces extend the main block. They add context, discovery, routing, or observation depending on the topic.
Claims registry
/claims.json
Registry of published claims, their scope, and their declarative status.
Entity graph
/entity-graph.jsonld
Descriptive graph of entities, identifiers, and relational anchor points.
Q-Layer in Markdown
/response-legitimacy.md
Canonical surface for response legitimacy, clarification, and legitimate non-response.
Authority conflict governance: advanced interpretive arbitration
Authority conflicts appear when several sources seem relevant, but do not speak with the same force, the same temporal validity, or the same domain competence. In AI systems, those conflicts are often flattened into a single smooth answer. That is precisely what this framework is meant to prevent.
Operational definition
Authority conflict governance is the bounded arbitration framework that determines how competing authority signals should be compared, ordered, or left unresolved when they cannot be reconciled without overreach.
Typology of conflicts
Typical conflict classes include:
- current vs outdated authority;
- central vs peripheral authority;
- doctrinal vs derivative authority;
- competent but narrow authority vs broader but weaker authority;
- explicit silence on one surface vs noisy implication on another.
Arbitration rules (GCA-1 to GCA-10)
GCA-1: explicit hierarchy
No arbitration should happen without a declared source hierarchy.
GCA-2: temporality
More recent authority does not always win, but time must be explicit whenever validity depends on update status.
GCA-3: domain competence
A source may be strong in one domain and weak outside it. Competence must be bounded.
GCA-4: prohibition on normative extrapolation
The system must not convert descriptive compatibility into normative authority.
GCA-5: unresolved conflict remains unresolved
If the sources cannot be legitimately reconciled, the answer should expose the conflict rather than invent an arbitration outcome.
GCA-6: traceability of the decision
If arbitration occurs, the path must remain explainable.
GCA-7: derivative surfaces stay derivative
A repository, index, or validator may help, but it should not silently outrank the declared canon.
GCA-8: authority shift must be intentional
A change of leading authority requires explicit governance, not accidental discoverability.
GCA-9: contestability
High-impact arbitration should remain open to audit and correction.
GCA-10: abstention remains legitimate
Where the conflict is real and unresolved, non-response may be the most faithful outcome.
When this framework applies
This framework is relevant whenever an AI system must produce an answer that depends on more than one source of authority. The typical trigger is not a factual error but a structural ambiguity: two legitimate sources disagree, and the system lacks an explicit rule to decide which one governs the answer. In practice, this arises frequently in environments where a Q-Layer defines response conditions but the upstream authority signals remain unranked or underspecified.
The framework connects directly to the concept of authority boundary. Without a declared boundary, the system cannot distinguish a source that is competent from one that merely appears relevant. It also intersects with interpretive governance at the structural level: arbitration rules are not policy preferences but governance infrastructure. They determine whether a system can claim fidelity to its canon or is instead producing unchecked inference.
Organizations operating at Layer 3 of authority governance will find this framework indispensable. At that level, the environment has moved beyond basic canon declaration and must handle competing signals with traceable, auditable logic. Without advanced conflict governance, any claim to auditability collapses the moment two sources diverge on the same attribute.
The external authority control doctrine further reinforces this point: when authority flows from outside the canonical perimeter, the arbitration framework is the only mechanism that prevents silent overwriting of internal truth by external convenience.
Why this framework matters
Authority conflict is one of the main places where plausible language hides governance failure. A system that arbitrates too easily stops being interpretable.
Read also
- Q-Layer
- Canon vs inference
- Interpretation trace
- Legitimate non-response