Doctrinal note: this text is meant to be read through External Authority Control (EAC), the layer that qualifies the admissibility of external authorities in interpretive reconstruction. See EAC: minimum doctrinal decisions · EAC doctrine.
This page assembles the full “interpretive governance” series. The goal is not to repeat the canon, but to provide a reading map, navigation paths, and direct access to phenomena, rules of authority, mechanisms of proof, and operating environments.
Canonical register: /definitions/.
Series map
- Series A: observable phenomena
- Series B: applied doctrine
- Series C: proof and audit
- Series D: environments
- Series E: debt and sustainability
Series A: observable phenomena
- Interpretive invisibilization: when information exists but disappears from the response
- Interpretive collision: entity fusion and synthesis hallucinations
- Interpretive capture: signal saturation and the diversion of truth
- Interpretive inertia: why corrections do not “stick”
- State drift: when AI freezes an outdated state (price, inventory, policy)
- Interpretive smoothing: why AI standardizes your thinking
Series B: applied doctrine (canon, authority, non-response)
- Authority boundary: what AI can deduce, and what it must not infer
- Canonical silence and legitimate non-response: sometimes “I don’t know” is the correct output
- Authority conflict: what to do when two strong sources oppose one another
Series C: proof and audit
- Proof of fidelity: why a citation is no longer enough
- Interpretation trace: making a response auditable without exposing the black box
- Canon-output gap: measuring distortion rather than debating the “true”
- Interpretive observability: the minimum metrics to log
Series D: environments (open web, RAG, agentic systems)
- Open web vs closed environments: governance does not operate in the same way
- Reliable RAG: why governance is a problem of boundaries, not retrieval
- Agentic systems: why non-response becomes a safety rule
Series E: debt and sustainability
- Interpretive debt: how it accumulates without spectacular failure
- Interpretive sustainability: correction budgets and version discipline
- Version power: why correction must be versioned like software
Reading paths
1) Stabilizing a brand (open web)
Start with A, then B, then C. Finish with D (open web) and E (debt + versioning).
2) Building a genuinely reliable RAG
Start with B, then C, then D (RAG). Finish with E.
3) Agentic systems (safety)
Start with B (authority boundary, non-response), then D (agentic systems), then C (trace + observability).
Going further
- Canonical register: /definitions/
- Doctrine: /doctrine/
- Frameworks: /frameworks/
- Clarifications: /clarifications/
- External Authority Control (EAC): doctrine · minimum decisions
Recommended entry points by symptom
This series is most useful when approached from a concrete symptom.
- An entity is visible but poorly understood: go through Expertise, then Entity disambiguation and Semantic collision reduction.
- The site seems readable but outputs remain unstable: open Machine-first semantic architecture, then Machine-first is not enough: why governance files change the reading regime.
- A response cites but distorts: read Proof of fidelity: why citation is no longer enough, Interpretation trace, and Canon-output gap.
- Effects are visible but weakly observed: open Observations, Q-Ledger, and Q-Metrics.
What the series does not replace
This series does not replace the Site role, the Machine-first canon, or the AI use policy.
It should be read as an editorial deployment built on top of already published governance surfaces, not as an autonomous authority source that would overturn the canonical core.