Skip to content

Article

Freshness and AI citation stability

Freshness is not automatically better than stability. The correct question is whether the claim is time-sensitive, canonical, obsolete or still valid.

CollectionArticle
TypeArticle
Categoryrisque interpretatif
Published2026-05-13
Updated2026-05-13
Reading time3 min

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.

  1. 01Definitions canon
  2. 02Site context
  3. 03Public AI manifest
Canon and identity#01

Definitions canon

/canon.md

Canonical surface that fixes identity, roles, negations, and divergence rules.

Governs
Public identity, roles, and attributes that must not drift.
Bounds
Extrapolations, entity collisions, and abusive requalification.

Does not guarantee: A canonical surface reduces ambiguity; it does not guarantee faithful restitution on its own.

Context and versioning#02

Site context

/site-context.md

Notice that qualifies the nature of the site, its reference function, and its non-transactional limits.

Governs
Editorial framing, temporality, and the readability of explicit changes.
Bounds
Silent drifts and readings that assume stability without checking versions.

Does not guarantee: Versioning makes a gap auditable; it does not automatically correct outputs already in circulation.

Entrypoint#03

Public AI manifest

/ai-manifest.json

Structured inventory of the surfaces, registries, and modules that extend the canonical entrypoint.

Governs
Access order across surfaces and initial precedence.
Bounds
Free readings that bypass the canon or the published order.

Does not guarantee: This surface publishes a reading order; it does not force execution or obedience.

Evidence layer

Probative surfaces brought into scope by this page

This page does more than point to governance files. It is also anchored to surfaces that make observation, traceability, fidelity, and audit more reconstructible. Their order below makes the minimal evidence chain explicit.

  1. 01
    Canon and scopeDefinitions canon
  2. 02
    Weak observationQ-Ledger
Canonical foundation#01

Definitions canon

/canon.md

Opposable base for identity, scope, roles, and negations that must survive synthesis.

Makes provable
The reference corpus against which fidelity can be evaluated.
Does not prove
Neither that a system already consults it nor that an observed response stays faithful to it.
Use when
Before any observation, test, audit, or correction.
Observation ledger#02

Q-Ledger

/.well-known/q-ledger.json

Public ledger of inferred sessions that makes some observed consultations and sequences visible.

Makes provable
That a behavior was observed as weak, dated, contextualized trace evidence.
Does not prove
Neither actor identity, system obedience, nor strong proof of activation.
Use when
When it is necessary to distinguish descriptive observation from strong attestation.

Freshness is not automatically better than stability. The correct question is whether the claim is time-sensitive, canonical, obsolete or still valid.

AI citation discussions often treat freshness as a universal advantage. That is too simple. Some claims require current data. Others require stable definitions. Some pages should change frequently. Others should remain intentionally stable so that systems can identify a durable source of meaning.

The risk is to update everything without governing what changed.

Four freshness regimes

A citation audit should classify claims by freshness regime.

RegimeExampleCitation risk
Time-sensitivepricing, schedule, regulation, availabilityoutdated citation
Versionedpolicy, methodology, specificationwrong version cited
Evergreendefinition, doctrine, stable conceptunnecessary churn weakens stability
Transitionalrebrand, migration, service changeold and new states collide

The same update strategy cannot govern all four regimes.

Freshness without scope creates drift

Updating a page can help if the previous state is obsolete. But updates can also create ambiguity if they do not declare what changed, what remains valid, and which older claims are no longer authoritative.

For AI-mediated systems, this matters because older content can remain retrievable through search indexes, third-party summaries, training memory, caches or citations. A new page does not automatically erase an old interpretation.

This is why known-source risk and phantom citation are linked to freshness. The system may believe it knows a source or URL even when the current site has moved on.

Citation stability is the stronger metric

Freshness asks whether the source is current. Citation stability asks whether the right source keeps governing the right claim across time, systems and prompt variants.

A page that is cited once after an update has not proven stability. A page that is repeatedly cited for the right claim, with the right role and without substitution, has stronger evidence.

This is why citation persistence and citation fidelity should be measured together.

Practical correction

For time-sensitive claims, expose dates, update cadence, validity windows and canonical current pages. For versioned claims, expose version numbers, effective dates and supersession routes. For evergreen doctrine, avoid unnecessary rewrites that weaken semantic continuity. For transitional states, create explicit bridges between old and new terminology.

A good page does not simply say “updated.” It says what the update governs.

Governance implication

Freshness becomes useful only when it is connected to source hierarchy. Without hierarchy, a system may cite the newest weak source instead of the strongest governing source. With hierarchy, the system has a clearer route: current operational claims go to current pages, definitions go to stable definitions, proof claims go to proof artifacts, and obsolete states are marked as such.

The goal is not maximal novelty. The goal is controlled temporal authority.