Interpretive debt
Status: canonical definition (lexical).
This page normatively defines the concept of interpretive debt within the interpretive governance doctrine framework. It serves to reduce ambiguity by declaring a stable and enforceable conceptual perimeter.
Back to registry: Definitions and canonical concepts.
Canonical definition
Interpretive debt: cumulative liability produced when approximations (on high-impact information) are repeated, reformulated, and stabilized by automated interpretation systems, until they become a default representation that is difficult to displace, making any subsequent correction more costly than an initial clarification.
Scope
Interpretive debt does not describe a “one-time error”. It describes an inertia that installs through multi-context circulation. It is particularly critical when it touches sensitive zones (information that structures classification, comparison, recommendation, or exclusion).
- It can exist even if marketing indicators (traffic, impressions, positions) remain stable.
- It increases correction cost, because the required action is no longer a simple precision, but an unanchoring.
- It is compatible with an appearance of coherence: an approximation can be “stable” without being legitimate.
Formation mechanism
- Initial approximation: a plausible, slightly encompassing or imprecise formulation appears.
- Repetition: the formulation is picked up by other contexts (syntheses, summaries, responses, comparisons).
- Stabilization: it becomes the dominant representation, cited and reused without verification.
- Rigidification: the cost of correction increases. The representation resists displacement (inertia).
- Extinction or regression: either the debt is resolved through governed correction, or it reappears (remanence).
Sensitive zones (where debt is most dangerous)
- Identity: who you are, what you do, what you do not do.
- Offering: what you offer, under what conditions, what you do not offer.
- Positioning: your sector, your competitors, your differentiators.
- Responsibilities: legal, regulatory, contractual implications.
- Recommendations: what AI suggests to users based on your representation.
Measurement signals
- Canon-output gap: divergence between what is declared and what AI produces.
- Compliance drift: progressive increase of gap without canon change.
- Remanence: reappearance of an old representation after correction.
- Trail: partial correction coexisting with persistent old interpretation.
- Identity incidents: collisions, contaminations, captures.
