Executive summary
An organisation can manage knowledge deliberately and still have a specific reliance question worth examining. Is the interpretation behind a consequential decision current? Can another qualified person explain an exception? Does an AI-generated summary preserve the conditions attached to its sources? Is the existing assessment well supported, or does it need qualification?
This paper proposes a place for ArtiProof within an organisational knowledge-management environment: a bounded, evidence-reusing review of whether particular knowledge is fit to be relied upon.
ISO 30401 provides requirements and guidance for a knowledge-management system. ArtiProof offers a proposed structure for reviewing a selected knowledge dependency inside that environment. This is a difference in emphasis and scope, not a claim that ISO 30401 excludes assurance, evidence, review or improvement. [1]
The ArtiProof review can confirm a supported conclusion, qualify it, challenge it or withhold judgement because evidence is insufficient. It need not invent another gap. Its distinct benefit must be tested against ordinary competent peer review, with comparable expertise, evidence access and effort.
The proposed relationship is therefore an existing knowledge-management system, with an optional knowledge-fitness review where it earns its place. It is not a merged standard, a second management system or a route to automatic conformity.
1. When a well-kept knowledge base is not the whole answer
Consider a hypothetical organisation that handles complex applications. Its knowledge base is organised, owners are identified and staff can retrieve approved procedures. A new assistant makes it easier to find the relevant material and prepare a recommendation.
One application exposes a conflict. A formally approved procedure reflects an earlier interpretation, while a later decision changes how a narrow exception should be handled. Both sources are authentic. The assistant gives a fluent answer, but it does not mention the exception.
The relevant issue is not whether the organisation “does knowledge management”. It is whether the knowledge supporting this particular recommendation is fit for this particular use.
A review might show that the later decision is controlling and that the recommendation needs correction. It might instead establish that the exception does not apply and confirm the existing interpretation. Or the available evidence might leave the question unresolved.
Each is a legitimate outcome. What matters is the relationship between the evidence, the intended reliance and the conclusion—not whether the review produces a favourable verdict or discovers something new.
This is the question ArtiProof proposes to structure. The example illustrates an application; it is not a reported pilot or evidence of a common defect in ISO 30401 implementations.
2. A management system and a bounded review
ISO describes ISO 30401 as a standard for establishing and improving a knowledge-management system, applicable across organisational types and sizes. That makes it a direct subject-matter neighbour for ArtiProof. [1]
The relationship should nevertheless be stated carefully. Knowledge management is not merely storage, and a management system is not inherently uncritical. An organisation may already have strong review, challenge, learning and evidence practices. The proposed ArtiProof profile must begin by finding out what those practices already accomplish.
ArtiProof’s object is narrower: the fitness of consequential knowledge for a defined use. The scope might be one decision, one professional activity, one service or a dependency crossing organisational boundaries. It should widen only when necessary to understand a material dependency, not because wider scope is a higher maturity level.
The operating arrangement is straightforward. The existing knowledge-management system supplies its objectives, responsibilities, practices and evidence. A selected ArtiProof review examines a particular reliance question. The resulting conclusion and any justified action return to the organisation’s existing decision and improvement processes.
The institution does not need a second committee, repository or improvement programme merely to use that structure. If an existing review already supplies the necessary evidence and challenge, the useful action may be to recognise it rather than repeat it.
3. Three knowledge roles, five separate questions
ArtiProof distinguishes Source Knowledge, Craft Knowledge and Generated Knowledge. These categories describe roles in a context, not three sealed compartments or three levels of truth. [3]
Source Knowledge provides an identifiable reference: an approved interpretation, a specification, a policy, a decision or a record. Its origin may be clear while its applicability remains uncertain.
Craft Knowledge is the practical understanding developed through experience and professional judgement. It may explain why an exception matters, how two sources interact or when routine processing should stop. Respecting this knowledge does not justify leaving an avoidable single-person dependency unmanaged.
Generated Knowledge is produced or materially transformed by a system: a summary, classification, inference or recommendation. An output can become a source for later work, but that change in role does not establish authority. Nor does human origin establish correctness.
The review examines five properties separately. Integrity concerns completeness and inappropriate alteration or degradation. Provenance concerns origin and development. Authority concerns the basis for reliance in the intended use. Understandability concerns the ability to interpret and appropriately challenge the knowledge. Resilience concerns preserving or recovering the knowledge-dependent capability.
The five properties do not combine into an organisational score. An authentic but superseded source can have strong provenance and weak suitability for reliance. A correct interpretation understood by one person can present a resilience problem. Excellent documentation cannot cancel out a material inability to interpret it.
These distinctions provide review prompts. Their existence is not evidence that ArtiProof produces better judgements than an experienced reviewer using another approach.
4. Start with the evidence already there
The proposed profile begins with evidence reuse. The organisation may already possess process descriptions, source registers, competence information, decision records, review findings, training material and handover demonstrations.
The task is to connect that material to the actual conclusion under review. A training certificate may support a claim that training occurred, but not establish that an unusual case can now be handled. A source register may identify ownership without proving that a particular interpretation is current. A successful demonstration may support capability under its tested conditions without proving performance in every context.
Reuse therefore includes a validity check: scope, relevance, age, material changes, ownership, reliability and the question the evidence actually answers.
Where a material uncertainty remains, an additional walkthrough, explanation, source comparison or bounded demonstration may be justified. The additional work should have a reason and a stopping condition. “More evidence” is not automatically better when it adds nothing to the decision.
A candidate Knowledge Matrix can make relationships visible between an activity, its required knowledge, relevant people, sources and systems. A candidate one-page view can summarise the conclusion and link to supporting evidence. Neither is a compulsory new inventory, a generic compliance score or a demonstrated improvement in efficiency. Both remain implementation patterns whose usefulness should be tested.
5. A proposed stale-knowledge challenge
A practical experiment for an ISO 30401 environment is a stale-knowledge challenge. This is an ArtiProof candidate practice, not an ISO-prescribed test or a new requirement of the standard.
Use a safe, synthetic or appropriately sanitised case in which authentic-looking material includes an obsolete instruction, a later qualification or an unresolved inconsistency. Specify the intended reliance question and the authoritative evidence available to reviewers before they begin.
Observe whether the reviewer notices the relevant difference, determines its significance and avoids treating a recent summary as a substitute for its controlling sources. The challenge should also include sound cases. Otherwise, reviewers learn that every exercise is a trap and may reject valid knowledge indiscriminately.
The original application example can be tested in three forms. In the first, the later decision genuinely changes the answer. In the second, it does not apply to the case, so confirmation is appropriate. In the third, the decisive evidence is unavailable, so withholding judgement is preferable to a confident guess.
Record the outcome, evidence used, assumptions, requests for clarification and effort. Do not reward a reviewer merely for disagreeing with the initial answer. Both false confirmation and unsupported challenge are errors.
The resulting observation concerns that exercise. A successful challenge is not ISO conformity evidence by default, and one successful exercise does not establish ArtiProof’s general validity.
6. Review conditions matter as much as the review label
A credible review record should make clear who reviewed what, for which use, with what competence and evidence access. Prior involvement in creating the assessment, conflicts and organisational relationships should remain visible.
Calling a reviewer “external” does not automatically establish independence. A separate department can still rely on the same selectively prepared evidence. A second AI response can repeat the same unsupported source. Conversely, a knowledgeable internal reviewer can provide useful challenge when the relationship and limitations are stated honestly.
The profile therefore proposes a compact description of review conditions rather than an unsupported badge of independence. It should distinguish evidence examined from evidence merely mentioned and identify material access restrictions.
Evidence absence also requires care. A record never created is different from a record that exists but was unavailable to the reviewer, was withheld or was no longer retained. Where the reason is unknown, it remains unknown. An explanation for the absence may improve transparency without supplying the substantive evidence needed for confirmation.
The conclusion should identify scope, intended use, period, material limitations and conditions for reconsideration. Confirm, qualify, challenge and insufficient evidence are review outcomes, not certification grades.
7. Human judgement and recovery
Knowledge assurance should not become an attempt to proceduralise every act of expertise. Some tasks require judgement precisely because their context cannot be fully specified in advance.
The relevant question is whether the organisation can support that judgement with appropriate competence, sources, decision rights, challenge and transfer arrangements. These conditions can be examined without assuming that every expert decision must be reduced to a rule.
Recovery also goes beyond retaining files. In a hypothetical supplier transition, the receiving team may obtain every database field yet lack the understanding needed to interpret an unusual value. A handover can therefore include a bounded demonstration that the receiving party can explain and handle a consequential case.
The objective is sufficient capability, not total self-sufficiency. Another qualified person, a different provider or an alternative tool may provide a sound recovery route. The appropriate arrangement depends on the intended use, consequences and acceptable interruption.
This is where the three knowledge roles are useful together. A preserved source, transferable craft understanding and appropriately controlled generated support may jointly sustain capability. None must be treated as a universal substitute for the others.
8. Proving value without inventing a gap
ArtiProof should not justify itself by portraying competent knowledge-management practice as inadequate before examining it.
A fair study would compare ordinary competent peer review with ArtiProof-guided peer review of the same kind of bounded assessment. Where feasible, a competent self-assessment condition provides a third reference point. Relevant expertise, access and task conditions should be matched; time differences should be controlled or reported.
Use cases whose expected conclusions are defensible independently of the method being tested: well-supported assessments, concealed material defects and insufficient-evidence cases. Freeze the case and outcome rules before execution. Record reviewers’ initial judgements before discussion, because agreement achieved through calibration is not evidence of independent initial agreement.
Relevant observations include warranted confirmation, false confirmation, justified qualification or challenge, correct withholding of judgement, evidence-to-conclusion traceability, clarification requests and review effort. Subjective confidence alone is not a benefit.
The review may have practical value without identifying a novel problem. However, if ordinary peer review reaches equally well-supported conclusions at equal or lower burden, the study has not demonstrated an ArtiProof-specific advantage. This is an acceptable outcome, not a reason to invent additional review machinery.
Peer review of an organisational assessment using ArtiProof must also be distinguished from independent review or validation of ArtiProof itself.
9. Edition-aware interoperability
The external baseline matters. At the source check for this release, ISO 30401:2018 remains the published edition, with Amendment 1:2022 and Amendment 2:2024. ISO/DIS 30401, Edition 2, remains a draft in the enquiry phase. A draft is not a published replacement. [1–2]
This paper makes a conceptual interoperability proposal using the official public scope and publication-status information. It does not reproduce the standard or claim a completed clause-by-clause conformity mapping. The full licensed standard and amendments were not used to certify a detailed clause crosswalk for this publication.
A formal implementation mapping can be prepared against the applicable published edition now, provided that the licensed text, amendments and actual implementation are examined. There is no need to postpone all conceptual work until a new edition appears. Any discussion of draft changes must remain separately labelled and must not silently become a current requirement.
When a replacement edition is published, recheck the mapping and its evidence assumptions. Keep the ArtiProof core independent of external clause numbering; version the interface instead.
Conclusion: make the reliance decision examinable
The strongest relationship is not that one framework “manages” while the other alone “assures”. An effective knowledge-management system may already do both.
ArtiProof proposes a particular way to focus a review: identify the consequential knowledge, examine the evidence across five separate properties, disclose the conditions of review and reach a bounded conclusion about reliance.
That structure should support useful challenge without forcing disagreement, useful confirmation without encouraging complacency, and sufficient evidence without generating a second bureaucracy.
Its place within an ISO 30401 environment will be earned through use and comparison, not through the assertion that an integration exists.
Manage knowledge deliberately. Review consequential reliance explicitly. Keep the evidence strong enough to support the conclusion—and the conclusion no broader than the evidence.
Publication boundary and references
This independent ArtiProof paper is informative. It is not an ISO publication, endorsement, certification route or substitute for the applicable standard. Applying ArtiProof does not establish ISO 30401 conformity; ISO 30401 conformity does not automatically establish a particular ArtiProof conclusion.
The proposed profile is research-ready for structured investigation, not independently validated as an integration. ArtiProof is not yet broadly independently validated, and its first independent replication remains pending. Examples and tests in this paper are illustrative or proposed, not reported results.
[1] International Organization for Standardization. ISO 30401:2018 — Knowledge management systems — Requirements. Official scope, published status and amendment listing. https://www.iso.org/standard/68683.html
[2] International Organization for Standardization. ISO/DIS 30401 — Knowledge management systems — Requirements, Edition 2. Official draft status and lifecycle. https://www.iso.org/standard/89436.html
[3] ArtiProof. Keeping Organisational Knowledge Dependable in the Age of AI. Public Whitepaper 1.2 EN, companion document in this release. Current methodology: https://artiproof.org/method/
[4] ArtiProof. Current development and validation status. https://artiproof.org/status/
Source-status check: September 2026. The internal publication record identifies the exact release check and the project decisions used. Publication version 1.1 does not designate a new normative method version.