AIM-001™

The SAFECHAIN™ Accountability Integrity Assessment Methodology™

Establishing the Methodology for Evidence-Based Assessment, Classification and Reassessment Across the SAFECHAIN™ AI1™–AI5™ Accountability Integrity Architecture

Framework Reference: AIM-001™
Framework Type: Assessment Methodology
Parent Framework: ACCOUNTABILITY-001™ — The SAFECHAIN™ Governance Answerability, Consequence & Institutional Accountability Framework™
Classification Architecture: AI1™–AI5™
Framework Series: SAFECHAIN™ Accountability Integrity Series
Author: Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA
Founder — SAFECHAIN™
Version: 1.0
Year: 2026

1. Purpose

The SAFECHAIN™ Accountability Integrity Assessment Methodology™ (AIM-001™) establishes the formal methodology for assessing the effectiveness, integrity and reliability of institutional accountability and determining the appropriate classification within the SAFECHAIN™ AI1™–AI5™ Accountability Integrity Architecture.

It provides a structured process for determining whether an organisation, institution, service, function, programme, decision pathway or defined governance system demonstrates:

AI1™ — Effective Accountability

AI2™ — Effective with Improvement

AI3™ — Material Accountability Gap

AI4™ — Serious Accountability Failure

AI5™ — Systemic Accountability Breakdown

AIM-001™ exists to ensure that classification is not based upon institutional reputation, policy existence, subjective confidence, isolated metrics or unsupported assertion.

Classification must follow evidence.

The methodology therefore examines both:

what an institution says its accountability system does

and

what the evidence demonstrates that system actually does in practice.

The central AIM-001™ principle is:

Accountability integrity should be assessed through evidence of institutional behaviour, decision-making, answerability, challenge, consequence, remediation and learning — not through the existence of governance structures alone.

The assessment pathway is:

Define → Scope → Preserve → Gather → Test → Corroborate → Analyse → Challenge → Classify → Report → Remediate → Verify → Reassess

2. Relationship with the Accountability Integrity Architecture

AIM-001™ operationalises the assessment methodology supporting:

• ACCOUNTABILITY-001™;

• AI1™;

• AI2™;

• AI3™;

• AI4™;

• AI5™.

The AI1™–AI5™ frameworks define the accountability conditions.

AIM-001™ determines how those conditions are assessed.

3. The Five Accountability Integrity Classifications

AI1™ — Effective Accountability

Accountability mechanisms are demonstrably effective, evidence-based, accessible, independent where necessary and capable of identifying responsibility, requiring answerability, applying consequence, correcting failure and preventing recurrence.

AI2™ — Effective with Improvement

Core accountability remains effective, but identifiable and bounded weaknesses require structured improvement.

AI3™ — Material Accountability Gap

One or more material weaknesses significantly reduce confidence that accountability will operate effectively within the assessed scope.

AI4™ — Serious Accountability Failure

Serious failure involving harm, compromised evidence, obstructed challenge, failed remediation, inadequate consequence, leadership failure or significant safeguarding concern substantially undermines accountability.

AI5™ — Systemic Accountability Breakdown

Accountability mechanisms are structurally, culturally or operationally incapable of reliably identifying, challenging, correcting or preventing serious failure across the affected system.

4. The SAFECHAIN™ Assessment Integrity Principle™

The purpose of assessment is not to achieve the most favourable classification available. It is to identify the classification most accurately supported by the evidence.

5. Assessment Objects

AIM-001™ may be applied to:

• An entire organisation;

• a governance function;

• a safeguarding system;

• a department;

• a service;

• a programme;

• a decision pathway;

• a complaints system;

• an assurance function;

• a multi-agency arrangement;

• a contractor ecosystem;

• a digital or AI-enabled decision system;

• another clearly defined accountability environment.

6. The SAFECHAIN™ Scope Integrity Principle™

An accountability classification is meaningful only when the scope to which it applies is clearly defined.

An AI classification should never be presented as applying to an entire organisation where only a limited process has been assessed unless the evidence justifies that wider conclusion.

7. Assessment Scope Statement™

Every AIM-001™ assessment should begin with a SAFECHAIN™ Assessment Scope Statement™.

It should identify:

• Assessment subject;

• organisational boundary;

• functions included;

• functions excluded;

• assessment period;

• locations where relevant;

• decision pathways;

• stakeholder groups;

• safeguarding relevance;

• technology or AI systems where applicable;

• known limitations.

8. Assessment Questions

The assessment should establish:

Who possesses authority?

Who owns responsibility?

How are decisions made?

How are decisions recorded?

Can decision-makers explain their reasoning?

Can evidence be tested?

Can challenge reach someone capable of changing the outcome?

Are conflicts managed?

Can failure produce consequence?

Can harm be identified and addressed?

Does remediation correct root cause?

Can oversight see material problems?

Does institutional learning prevent recurrence?

9. The SAFECHAIN™ Accountability Integrity Assessment Domains™

AIM-001™ establishes twelve core assessment domains.

AID1 — Authority Integrity

Is authority legitimate, documented, current and exercised within scope?

AID2 — Responsibility Integrity

Can responsibility for decisions, duties and outcomes be reliably traced?

AID3 — Answerability Integrity

Can relevant decision-makers provide substantive explanations?

AID4 — Evidence Integrity

Can consequential decisions and events be reconstructed and tested?

AID5 — Independence & Impartiality

Are conflicts controlled and independent scrutiny available?

AID6 — Challenge Integrity

Can concerns, dissent and contrary evidence meaningfully affect decisions?

AID7 — Safeguarding Integrity

Can the institution recognise, prevent, escalate and respond to harm?

AID8 — Consequence Integrity

Does established failure result in proportionate response?

AID9 — Remediation Integrity

Does corrective action address root cause and operate in practice?

AID10 — Oversight Integrity

Can governing bodies detect, challenge and require correction?

AID11 — Learning Integrity

Does the institution translate findings into demonstrable change?

AID12 — Accountability Culture

Does institutional culture support truth, ownership, challenge and correction?

10. Assessment Phases

AIM-001™ establishes nine principal assessment phases:

Phase 1 — Initiation

Phase 2 — Scoping

Phase 3 — Evidence Preservation

Phase 4 — Evidence Collection

Phase 5 — Evidence Testing

Phase 6 — Domain Assessment

Phase 7 — Classification

Phase 8 — Reporting & Challenge

Phase 9 — Verification & Reassessment

11. Phase 1 — Initiation

The assessment should identify:

• Why the assessment is being undertaken;

• who commissioned it;

• what authority supports it;

• who will receive findings;

• whether safeguarding concerns exist;

• whether urgent controls are required.

12. Assessment Trigger

An AIM-001™ assessment may arise from:

• Routine governance review;

• certification or accreditation;

• safeguarding concern;

• serious incident;

• audit finding;

• complaint pattern;

• whistleblowing;

• regulatory finding;

• leadership request;

• board concern;

• failed remediation;

• evidence of recurring accountability weakness.

13. SAFECHAIN™ Risk-Led Assessment Trigger™

The greater the credible indication of harm, evidence compromise, leadership involvement, safeguarding failure or obstruction of challenge, the stronger the requirement for timely and sufficiently independent assessment.

14. Phase 2 — Scoping

The assessor should define exactly what is being assessed.

Scope should not be:

• Artificially narrow;

• unnecessarily broad;

• designed around a preferred outcome;

• altered to exclude inconvenient evidence.

15. SAFECHAIN™ Evidence-Led Scope Principle™

Assessment scope should be capable of expanding where credible evidence demonstrates that the accountability issue extends beyond the original boundary.

16. Assessment Period

Every assessment should identify a defined assessment period.

Historical evidence may be examined where necessary to understand:

• Recurrence;

• previous warnings;

• prior remediation;

• leadership knowledge;

• systemic patterns.

17. Phase 3 — Evidence Preservation

Where material or serious accountability concerns exist, relevant evidence should be preserved before detailed assessment proceeds.

This may include:

• Emails;

• decision records;

• governance papers;

• complaints;

• audit trails;

• system logs;

• safeguarding records;

• meeting records;

• policies;

• investigation records;

• remediation records.

18. SAFECHAIN™ Assessment Preservation Principle™

An accountability assessment cannot reliably determine what happened if material evidence is allowed to disappear while the assessment is underway.

19. Phase 4 — Evidence Collection

AIM-001™ requires evidence from multiple sources where proportionate.

Potential sources include:

• Policies;

• procedures;

• delegations;

• decision records;

• correspondence;

• complaints;

• incident records;

• safeguarding material;

• audit reports;

• assurance reports;

• board papers;

• performance data;

• interviews;

• stakeholder testimony;

• regulatory findings;

• independent reports;

• system records;

• remediation evidence.

20. SAFECHAIN™ Evidence Triangulation Principle™

Material accountability conclusions should, wherever reasonably possible, be tested across more than one evidential source rather than relying exclusively upon institutional self-reporting.

21. Evidence Hierarchy

Evidence should be evaluated according to:

• Reliability;

• contemporaneity;

• independence;

• completeness;

• relevance;

• corroboration;

• authenticity.

No evidence type should automatically determine the outcome without context.

22. Primary Evidence

Primary evidence may include:

• Original decision records;

• contemporaneous emails;

• system logs;

• meeting minutes;

• signed records;

• contemporaneous complaints;

• source data;

• original safeguarding records.

23. Secondary Evidence

Secondary evidence may include:

• Later summaries;

• retrospective accounts;

• management reports;

• reconstructed chronologies;

• analysis based on primary material.

24. Lived and Stakeholder Evidence

Evidence from people affected by institutional processes should be treated as potentially material governance evidence.

It should not automatically be dismissed because it is:

• Experiential;

• qualitative;

• critical of the institution;

• inconsistent with management assurance.

25. SAFECHAIN™ Lived Evidence Integrity Principle™

Where accountability is intended to operate for people affected by institutional power, their evidence about how that accountability operates in practice may be essential to assessing governance effectiveness.

26. Missing Evidence

Missing evidence is itself relevant.

The assessor should establish:

• What should exist;

• whether it exists;

• why it is missing;

• who controlled it;

• whether retention requirements applied;

• what effect its absence has upon assurance.

27. SAFECHAIN™ Missing Evidence Assessment Rule™

Where evidence necessary to demonstrate effective accountability should reasonably exist but cannot be produced, the absence should reduce assurance rather than automatically favour the institution.

28. Contradictory Evidence

Contradictions should be identified rather than averaged away.

The assessor should ask:

Which evidence is contemporaneous?

Which evidence is independently corroborated?

Which source had direct knowledge?

Which source had an interest in the outcome?

Can the contradiction be resolved?

29. SAFECHAIN™ Contradiction Resolution Record™

Material contradictions should be recorded together with:

• Competing evidence;

• assessment;

• reasoning;

• unresolved uncertainty.

30. Phase 5 — Evidence Testing

Evidence should be tested against:

• Policy;

• actual practice;

• decision records;

• stakeholder experience;

• outcomes;

• applicable governance obligations;

• previous findings;

• remediation claims.

31. The SAFECHAIN™ Policy-to-Practice Test™

Ask:

What does policy require?

What happened in practice?

What evidence demonstrates that?

What explains any difference?

What accountability consequence follows?

32. Practice Prevails

A policy should not be treated as evidence of effectiveness where operational evidence demonstrates otherwise.

33. SAFECHAIN™ Operational Reality Principle™

For accountability assessment purposes, demonstrated institutional behaviour carries greater assurance value than an untested statement of intended practice.

34. Sampling

Sampling may be used where assessing large systems.

Sampling methodology should consider:

• Risk;

• materiality;

• safeguarding;

• complaint history;

• geography;

• service type;

• leadership;

• vulnerable groups;

• known failure patterns.

35. SAFECHAIN™ Risk-Based Sampling Principle™

Sampling should be designed to test where accountability is most likely to fail as well as where it is expected to succeed.

36. Random Sampling

Random samples may help assess general control operation.

37. Targeted Sampling

Targeted sampling should be used where specific risks exist.

Examples include:

• Serious complaints;

• safeguarding incidents;

• disputed decisions;

• repeated failures;

• previous remediation.

38. Adverse Case Sampling™

AIM-001™ establishes SAFECHAIN™ Adverse Case Sampling™.

This deliberately examines cases where the accountability system was placed under pressure.

The purpose is to determine whether accountability works when it is most needed.

39. SAFECHAIN™ Stress-of-Accountability Principle™

An accountability system should not be judged solely by routine cases; its integrity is most clearly tested when evidence is disputed, harm is alleged, authority is challenged or institutional interests are engaged.

40. Interviews

Interviews may be conducted with:

• Leadership;

• board members;

• staff;

• safeguarding personnel;

• auditors;

• investigators;

• complainants;

• service users;

• affected stakeholders;

• third parties.

41. Interview Integrity

Interview evidence should be recorded accurately and distinguished from verified fact.

42. SAFECHAIN™ Protected Participation Principle™

Where appropriate, assessment arrangements should protect participants from improper retaliation or disadvantage.

43. Phase 6 — Domain Assessment

Each of the twelve domains should be assessed separately before determining the overall classification.

44. SAFECHAIN™ Domain Integrity Rating™

Each domain may receive one of five conditions:

D1 — Effective

D2 — Effective with Improvement

D3 — Material Gap

D4 — Serious Failure

D5 — Systemic Breakdown

These domain ratings inform, but do not mechanically determine, the overall AI classification.

45. Authority Assessment

Test:

• Source of authority;

• delegation;

• scope;

• expiry;

• decision rights;

• unauthorised influence;

• shadow decision-making.

46. Responsibility Assessment

Test:

• Named ownership;

• duty allocation;

• escalation responsibility;

• supervisory responsibility;

• executive ownership;

• board responsibility.

47. Answerability Assessment

Test whether:

• Decision-makers are identifiable;

• reasoning exists;

• explanations address the evidence;

• questions receive substantive answers;

• decisions can be reconstructed.

48. Evidence Assessment

Test:

• Completeness;

• preservation;

• reliability;

• traceability;

• accessibility;

• consistency;

• auditability.

49. Independence Assessment

Test:

• Conflicts;

• self-review;

• recusal;

• investigator independence;

• scope control;

• reporting freedom.

50. Challenge Assessment

Test:

• Complaint access;

• escalation;

• appeals;

• dissent;

• whistleblowing;

• retaliation;

• ability to introduce contrary evidence.

51. Safeguarding Assessment

Test:

• Recognition;

• escalation;

• protection;

• vulnerability;

• participation;

• response to harm;

• learning.

52. Consequence Assessment

Test whether established failure produces:

• Corrective action;

• accountability;

• authority restriction where appropriate;

• disciplinary or contractual processes where applicable;

• leadership response;

• systemic change.

53. Remediation Assessment

Test:

• Root cause;

• ownership;

• deadlines;

• implementation;

• effectiveness;

• recurrence;

• verification.

54. Oversight Assessment

Test:

• Board visibility;

• reporting quality;

• independent challenge;

• assurance;

• intervention;

• closure scrutiny.

55. Learning Assessment

Test whether:

• Lessons are identified;

• actions follow;

• change is implemented;

• recurrence is monitored;

• learning is shared.

56. Culture Assessment

Test whether institutional behaviour supports:

• Candour;

• ownership;

• challenge;

• correction;

• evidence;

• safeguarding;

• learning.

57. Cross-Domain Analysis

Accountability weaknesses should not always be considered independently.

A weakness in one domain may create failure elsewhere.

For example:

Evidence failure → weak answerability → ineffective challenge → no consequence → failed remediation.

58. SAFECHAIN™ Accountability Failure Chain™

AIM-001™ requires assessors to identify where failures create or reinforce other failures across the accountability chain.

59. Cumulative Assessment

Several individually limited weaknesses may collectively justify a higher classification.

60. SAFECHAIN™ Cumulative Accountability Principle™

Overall accountability integrity should reflect the combined governance effect of identified weaknesses rather than treating every weakness as an isolated event.

61. Materiality

Materiality should consider:

• Harm;

• rights;

• safeguarding;

• scale;

• duration;

• recurrence;

• leadership knowledge;

• vulnerability;

• evidence integrity;

• correctability.

62. SAFECHAIN™ Accountability Materiality Assessment™

AIM-001™ assesses:

AM1 — Impact

AM2 — Reach

AM3 — Duration

AM4 — Recurrence

AM5 — Vulnerability

AM6 — Evidence Integrity

AM7 — Leadership Knowledge

AM8 — Challenge Effectiveness

AM9 — Remediation Effectiveness

AM10 — Institutional Significance

63. Severity

Severity should not be determined by financial impact alone.

A failure affecting:

• Safety;

• rights;

• dignity;

• fairness;

• evidence;

• access to remedy;

may be highly significant even where monetary impact is limited.

64. Recurrence

Repeated failure is a critical assessment indicator.

65. SAFECHAIN™ Recurrence Escalation Principle™

Where substantially similar accountability weaknesses recur after the institution has had a reasonable opportunity to correct them, recurrence should increase the level of governance concern.

66. Previous Remediation

Assessors should examine:

Was the issue previously identified?

What was promised?

Was it implemented?

Was it verified?

Did the problem recur?

67. Leadership Knowledge

Leadership knowledge may materially affect classification.

Assess:

• Actual knowledge;

• reasonable knowledge;

• warning signals;

• authority to intervene;

• action taken;

• failure to act.

68. Safeguarding Weighting

Safeguarding concerns require heightened attention.

69. SAFECHAIN™ Safeguarding Significance Principle™

Where accountability weakness materially increases the risk of serious harm to vulnerable people, classification should reflect the safeguarding significance of the failure rather than treating it as an ordinary process defect.

70. Phase 7 — Classification

Classification should follow completion of sufficient evidence analysis.

The assessor should determine which condition best represents the evidence:

AI1™

AI2™

AI3™

AI4™

or

AI5™

71. SAFECHAIN™ Classification Integrity Rule™

The classification should represent the actual accountability condition demonstrated by the evidence at the date of assessment, not the condition the institution expects to achieve after remediation.

72. Classification Is Not an Average

AIM-001™ does not require averaging all domain ratings.

A serious safeguarding or evidence-integrity failure may justify a higher overall classification despite stronger performance elsewhere.

73. SAFECHAIN™ Critical Domain Override™

A Critical Domain Override™ may apply where failure within a critical domain is sufficiently serious that averaging would materially understate accountability risk.

Potential critical domains include:

• Safeguarding;

• evidence integrity;

• independence;

• challenge;

• leadership accountability.

74. AI1™ Classification Threshold

AI1™ requires evidence that accountability is consistently effective across the assessed scope.

Minor observations may exist, but they should not materially weaken accountability.

75. AI2™ Classification Threshold

AI2™ applies where:

• Core accountability remains effective;

• weaknesses are bounded;

• risk is controlled;

• improvement is credible;

• no material accountability gap exists.

76. AI3™ Classification Threshold

AI3™ applies where:

• A material gap exists;

• accountability assurance is significantly reduced;

• formal remediation is required;

• ordinary improvement is insufficient.

77. AI4™ Classification Threshold

AI4™ applies where:

• Serious harm or risk exists;

• evidence integrity is seriously compromised;

• challenge is obstructed;

• serious safeguarding failure exists;

• remediation has materially failed;

• leadership or oversight failure is significant.

78. AI5™ Classification Threshold

AI5™ applies where:

• Failure is systemic;

• multiple accountability mechanisms are compromised;

• ordinary internal correction cannot be reliably trusted;

• structural reconstruction is required.

79. Classification Confidence

Every assessment should distinguish classification from confidence.

80. SAFECHAIN™ Assessment Confidence Scale™

AC1 — Preliminary

Evidence suggests a classification but material investigation remains outstanding.

AC2 — Reasonable

Sufficient evidence supports the classification, subject to identified limitations.

AC3 — Strong

Multiple reliable sources support the classification.

AC4 — Independently Corroborated

Material conclusions are supported by sufficiently independent corroboration.

81. Classification Confidence Is Not Severity

An AI5™ assessment may initially have AC1 confidence.

An AI2™ assessment may have AC4 confidence.

Severity and evidential confidence are separate.

82. Uncertainty

Assessment reports should state material uncertainty.

83. SAFECHAIN™ Uncertainty Integrity Principle™

Where evidence does not permit a reliable conclusion, uncertainty should be reported rather than replaced by unjustified certainty.

84. Precautionary Controls

Uncertainty does not necessarily prevent protective action.

Where serious harm is credibly indicated, proportionate interim controls may be appropriate while assessment continues.

85. Assessor Independence

The required level of independence should reflect severity and conflict risk.

86. SAFECHAIN™ Assessment Independence Scale™

IN1 — Operational Self-Assessment

Suitable for routine low-risk review.

IN2 — Internal Independent Assessment

Conducted by a function sufficiently separate from the activity assessed.

IN3 — Enhanced Independent Assessment

Greater organisational separation and formal conflict controls.

IN4 — External Independent Assessment

Used where internal independence cannot provide sufficient credibility.

87. Independence Escalation

Higher-severity matters should ordinarily require greater assessor independence.

AI4™ and AI5™ assessments should trigger explicit consideration of whether external independence is necessary.

88. Assessor Conflict Declaration™

Every formal assessment should record relevant:

• Financial interests;

• reporting relationships;

• previous involvement;

• professional relationships;

• personal conflicts;

• institutional dependencies.

89. SAFECHAIN™ Assessor Recusal Principle™

An assessor should not determine accountability integrity where a material conflict prevents credible independent judgement.

90. Competence

Assessors should possess competence appropriate to the scope.

Relevant competence may include:

• Governance;

• safeguarding;

• audit;

• investigations;

• law;

• risk;

• data;

• technology;

• organisational behaviour.

91. Multi-Disciplinary Assessment

Complex AI4™ and AI5™ assessments may require multi-disciplinary expertise.

92. Phase 8 — Reporting

The assessment report should clearly distinguish:

Evidence

Finding

Analysis

Classification

Recommendation

93. SAFECHAIN™ Finding Traceability Standard™

Every material finding should be capable of being traced to the evidence and reasoning supporting it.

94. Assessment Report Structure

A formal report should ordinarily include:

1. Executive Summary

2. Assessment Scope

3. Methodology

4. Evidence Base

5. Limitations

6. Domain Findings

7. Materiality Assessment

8. Safeguarding Assessment

9. Leadership & Oversight Findings

10. Overall Classification

11. Classification Confidence

12. Remediation Requirements

13. Verification Requirements

14. Reassessment Date

95. Findings

Potential findings may be classified as:

O — Observation

I — Improvement

M — Material Gap

S — Serious Failure

B — Systemic Breakdown

96. Findings Register

A SAFECHAIN™ Accountability Integrity Findings Register™ should record:

• Finding reference;

• domain;

• evidence;

• severity;

• materiality;

• responsible owner;

• required action;

• deadline;

• verification;

• status.

97. Right of Factual Challenge

The assessed institution should ordinarily be given an opportunity to identify:

• Factual inaccuracies;

• missing material evidence;

• misattribution;

• demonstrable methodological error.

98. SAFECHAIN™ Challenge-without-Control Principle™

The right to challenge an assessment should permit correction of genuine error without allowing the assessed institution to control, dilute or suppress independent findings.

99. Disagreement

Disagreement with a finding does not automatically invalidate it.

Material disagreement should be:

• Recorded;

• evaluated;

• resolved where possible;

• transparently retained where unresolved.

100. Management Response

The institution may provide a formal management response.

The response should remain distinguishable from the assessor's findings.

101. Publication and Communication

Where assessment results are communicated publicly, they should state:

• Scope;

• date;

• classification;

• confidence where appropriate;

• material limitations.

102. SAFECHAIN™ Classification Representation Principle™

No AI1™–AI5™ classification should be represented more broadly, more favourably or for longer than the evidence and assessment scope justify.

103. Classification Duration

Classification is time-specific.

It should not be treated as permanent.

104. Reassessment

Reassessment may be triggered by:

• Scheduled review;

• significant incident;

• failed remediation;

• new evidence;

• organisational restructuring;

• safeguarding concern;

• material regulatory finding;

• substantial improvement.

105. SAFECHAIN™ Classification Currency Principle™

An accountability classification remains credible only while there is sufficient evidence that the conditions supporting it continue to exist.

106. Phase 9 — Verification

Remediation should be verified before classification is improved.

107. SAFECHAIN™ Remediation Verification Test™

Ask:

Was the action completed?

Did the underlying condition change?

Does the control operate?

Has behaviour changed?

Has recurrence reduced?

Can improvement be evidenced independently?

108. Action Completion Is Not Effectiveness

A closed action is not necessarily a corrected accountability weakness.

109. SAFECHAIN™ Completion-versus-Correction Principle™

Remediation should be judged by whether the accountability condition has changed, not merely whether the administrative action has been marked complete.

110. Reclassification

Reclassification should require sufficient evidence.

Movement may occur:

AI5™ → AI4™, AI3™, AI2™ or AI1™

AI4™ → AI3™, AI2™ or AI1™

AI3™ → AI2™ or AI1™

AI2™ → AI1™

Deterioration may occur in the opposite direction.

111. No Mandatory Sequential Movement

Classification should follow evidence rather than a predetermined sequence.

112. SAFECHAIN™ Evidence-Led Reclassification Principle™

An institution should move to whichever accountability classification the current evidence supports, regardless of its previous classification.

113. Escalation During Assessment

Assessment should not wait for final reporting where evidence reveals urgent:

• Safeguarding risk;

• evidence destruction;

• retaliation;

• continuing serious harm;

• compromised decision-making.

114. SAFECHAIN™ Immediate Escalation Principle™

Assessment methodology should never become a reason to delay proportionate protective action where credible evidence demonstrates serious continuing risk.

115. Multi-Agency Assessment

Where several organisations contribute to an outcome, AIM-001™ should map:

• Authority;

• responsibility;

• information flows;

• safeguarding duties;

• escalation;

• oversight;

• remediation ownership.

116. SAFECHAIN™ Cross-Institution Accountability Assessment™

Each organisation should be assessed both for:

its own accountability

and

its contribution to collective accountability.

117. Third-Party Accountability

Outsourcing does not automatically transfer accountability.

Assess:

• Contractual responsibility;

• oversight;

• monitoring;

• escalation;

• evidence access;

• safeguarding;

• corrective authority.

118. Artificial Intelligence and Automated Systems

AIM-001™ applies to AI-enabled institutional decision-making.

Assessment should examine:

• Human accountability owner;

• deployment authority;

• data integrity;

• model validation;

• explainability;

• human review;

• challenge;

• monitoring;

• incident response;

• override capability.

119. SAFECHAIN™ Human Accountability Continuity Test™

Ask:

Can a named human authority ultimately explain, review and take responsibility for the institutional use of the automated system?

If not, a material accountability concern exists.

120. Algorithmic Evidence

AI-generated explanations should not automatically be treated as independent evidence of system integrity.

Underlying:

• Data;

• logs;

• decision pathways;

• validation;

• human controls;

should be examined where material.

121. Assessment Manipulation Risk

Assessors should remain alert to:

• Selective evidence;

• curated samples;

• coached interviews;

• delayed disclosure;

• changed records;

• restricted access;

• scope manipulation.

122. SAFECHAIN™ Assessment Gaming Principle™

Where an institution materially manipulates the conditions under which accountability is assessed, the manipulation itself becomes relevant evidence about accountability integrity.

123. Good Performance

AIM-001™ is not designed only to detect failure.

It should also recognise evidence of effective accountability.

Examples include:

• Clear ownership;

• documented authority;

• candid error correction;

• protected challenge;

• transparent reasoning;

• proportionate consequence;

• independently verified remediation;

• measurable learning.

124. SAFECHAIN™ Positive Accountability Evidence™

Evidence of effective governance should be recorded with the same discipline as evidence of failure.

125. No Presumption of Failure

Assessment should remain impartial.

126. SAFECHAIN™ Neutral Assessment Principle™

AIM-001™ should neither presume institutional effectiveness nor institutional failure; both must be demonstrated through evidence.

127. Assessment Quality Assurance

Formal assessments should be subject to proportionate quality assurance.

Quality assurance may examine:

• Scope;

• evidence sufficiency;

• reasoning;

• consistency;

• conflicts;

• classification;

• report clarity.

128. SAFECHAIN™ Assessment Quality Review™

A quality reviewer should ask:

Is the classification supported by the evidence?

Were contradictory facts addressed?

Was materiality properly assessed?

Were safeguarding issues weighted appropriately?

Was independence sufficient?

Can findings be traced?

129. Assessment Record

A complete assessment record should preserve:

• Scope;

• methodology;

• evidence index;

• interview records;

• working analysis;

• conflicts;

• findings;

• classification;

• quality review;

• final report;

• remediation;

• reassessment.

130. SAFECHAIN™ Accountability Integrity Assessment Record™

The assessment record provides the evidential audit trail for the classification.

131. Core Assessment Dashboard

A SAFECHAIN™ AIM-001 Accountability Integrity Dashboard™ may display:

Domain | Condition | Evidence Confidence | Materiality | Safeguarding | Trend | Required Action

The dashboard should support analysis rather than replace it.

132. Overall Assessment Logic

AIM-001™ uses four layers:

Layer 1 — Evidence

What happened?

Layer 2 — Domain Condition

Which accountability mechanism is affected?

Layer 3 — Materiality & Severity

How significant is the weakness?

Layer 4 — Classification

Which AI1™–AI5™ condition best represents the total evidence?

133. SAFECHAIN™ Four-Layer Assessment Model™

The model prevents premature classification by requiring assessors to move from:

Evidence → Domain → Materiality → Classification

rather than selecting a classification first and searching for evidence to support it.

134. The SAFECHAIN™ AIM-001 Assessment Integrity Test™

Before issuing an AI1™–AI5™ classification, ask:

1. Is the assessment scope clear?

2. Is the assessment period defined?

3. Is assessor authority established?

4. Are conflicts declared?

5. Is independence proportionate to risk?

6. Is the evidence base sufficient?

7. Was relevant evidence preserved?

8. Were primary records examined?

9. Was institutional self-reporting independently tested?

10. Were affected stakeholders appropriately considered?

11. Were material contradictions identified?

12. Were missing records examined?

13. Was policy compared with practice?

14. Was sampling sufficiently risk-based?

15. Were adverse cases tested?

16. Was authority assessed?

17. Was responsibility traced?

18. Was answerability tested?

19. Was evidence integrity assessed?

20. Was independence tested?

21. Was challenge effectiveness assessed?

22. Was safeguarding assessed?

23. Was consequence assessed?

24. Was remediation effectiveness assessed?

25. Was oversight tested?

26. Was institutional learning examined?

27. Was culture considered?

28. Were cumulative weaknesses assessed?

29. Was recurrence examined?

30. Was leadership knowledge tested?

31. Was materiality properly determined?

32. Were critical-domain failures considered?

33. Is the proposed classification supported by the actual evidence rather than planned improvement?

34. Is classification confidence separately recorded?

35. Were limitations and uncertainty disclosed?

36. Was factual challenge permitted without surrendering assessor independence?

37. Can every material finding be traced to evidence?

38. Has the assessment received appropriate quality assurance?

39. Are remediation and verification requirements clear?

40. If all policy statements, institutional assurances, reputation, intentions and planned reforms were removed from consideration, would the underlying evidence of actual authority, decisions, records, answerability, challenge, safeguarding, consequence, remediation, oversight and recurrence still support the proposed AI classification?

If yes, the classification has passed the central:

SAFECHAIN™ AIM-001 Assessment Integrity Test™

135. Assessment Outcomes

Implementation of AIM-001™ is intended to provide:

✓ Consistent AI1™–AI5™ classification
✓ Evidence-led governance assessment
✓ Clear assessment scope
✓ Stronger assessor independence
✓ Better evidence preservation
✓ Evidence triangulation
✓ Recognition of missing evidence
✓ Policy-to-practice testing
✓ Risk-based sampling
✓ Adverse case testing
✓ Stronger safeguarding assessment
✓ Better responsibility tracing
✓ Stronger answerability analysis
✓ Independent challenge testing
✓ Better materiality assessment
✓ Recognition of cumulative failure
✓ Leadership accountability analysis
✓ Clear classification confidence
✓ Transparent reporting
✓ Traceable findings
✓ Verified remediation
✓ Evidence-based reclassification
✓ Stronger institutional learning

136. Governing Statement

An accountability classification has value only if people can trust how it was reached.

A policy is not proof.

A committee is not proof.

An assurance statement is not proof.

A completed action plan is not proof.

A low number of complaints is not proof.

An organisation describing itself as accountable is not proof.

Accountability integrity must be demonstrated.

That requires evidence.

It requires the ability to trace authority.

To identify responsibility.

To reconstruct consequential decisions.

To test explanations.

To hear challenge.

To examine safeguarding.

To determine whether failure produces consequence.

To establish whether remediation actually works.

To determine whether oversight sees the truth.

And to ask whether the same failure happens again.

AIM-001™ therefore establishes a simple but demanding principle:

Do not classify the institution according to what its governance architecture promises. Classify it according to what the evidence demonstrates that architecture actually delivers.

The assessor must begin with the evidence rather than the desired conclusion.

The scope must be clear.

Contradictions must remain visible.

Missing evidence must matter.

Those affected must not disappear from the assessment.

Serious safeguarding concerns must not be averaged away.

Institutional reputation must not substitute for verification.

Planned reform must not be mistaken for present effectiveness.

And classification must change when the evidence changes.

The governing methodology is:

Define the scope. Preserve the evidence. Follow authority. Trace responsibility. Test the records. Hear challenge. Examine harm. Compare policy with practice. Investigate recurrence. Test leadership knowledge. Assess materiality. Classify the condition. Verify correction. Reassess the reality.

That is how the SAFECHAIN™ Accountability Integrity Assessment Methodology™ converts AI1™–AI5™ from a classification architecture into an evidence-based governance assessment system.

Copyright and Intellectual Property Notice

© 2026 Samantha Avril-Andreassen. All Rights Reserved.

AIM-001™ — The SAFECHAIN™ Accountability Integrity Assessment Methodology™ is an original governance assessment, accountability classification, evidence analysis, materiality, safeguarding, verification and institutional-assurance methodology developed and authored by Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA, Founder of SAFECHAIN™.

AIM-001™ forms part of the SAFECHAIN™ Accountability Integrity Series and provides the assessment methodology supporting ACCOUNTABILITY-001™ and the AI1™–AI5™ Accountability Integrity Classification Architecture.

The original expression, structure, architecture, arrangement, terminology, assessment methodology, domain model, assessment phases, sampling methodologies, classification logic, confidence methodology, principles, tests, registers, dashboards, verification mechanisms and associated materials contained within this publication constitute proprietary intellectual property.

This includes, where original to this methodology, the:

• AIM-001™ designation;

• SAFECHAIN™ Accountability Integrity Assessment Methodology™;

• SAFECHAIN™ Assessment Integrity Principle™;

• SAFECHAIN™ Scope Integrity Principle™;

• SAFECHAIN™ Assessment Scope Statement™;

• SAFECHAIN™ Accountability Integrity Assessment Domains™;

• SAFECHAIN™ Risk-Led Assessment Trigger™;

• SAFECHAIN™ Evidence-Led Scope Principle™;

• SAFECHAIN™ Assessment Preservation Principle™;

• SAFECHAIN™ Evidence Triangulation Principle™;

• SAFECHAIN™ Lived Evidence Integrity Principle™;

• SAFECHAIN™ Missing Evidence Assessment Rule™;

• SAFECHAIN™ Contradiction Resolution Record™;

• SAFECHAIN™ Policy-to-Practice Test™;

• SAFECHAIN™ Operational Reality Principle™;

• SAFECHAIN™ Risk-Based Sampling Principle™;

• SAFECHAIN™ Adverse Case Sampling™;

• SAFECHAIN™ Stress-of-Accountability Principle™;

• SAFECHAIN™ Protected Participation Principle™;

• SAFECHAIN™ Domain Integrity Rating™;

• SAFECHAIN™ Accountability Failure Chain™;

• SAFECHAIN™ Cumulative Accountability Principle™;

• SAFECHAIN™ Accountability Materiality Assessment™;

• SAFECHAIN™ Recurrence Escalation Principle™;

• SAFECHAIN™ Safeguarding Significance Principle™;

• SAFECHAIN™ Classification Integrity Rule™;

• SAFECHAIN™ Critical Domain Override™;

• SAFECHAIN™ Assessment Confidence Scale™;

• SAFECHAIN™ Uncertainty Integrity Principle™;

• SAFECHAIN™ Assessment Independence Scale™;

• SAFECHAIN™ Assessor Conflict Declaration™;

• SAFECHAIN™ Assessor Recusal Principle™;

• SAFECHAIN™ Finding Traceability Standard™;

• SAFECHAIN™ Accountability Integrity Findings Register™;

• SAFECHAIN™ Challenge-without-Control Principle™;

• SAFECHAIN™ Classification Representation Principle™;

• SAFECHAIN™ Classification Currency Principle™;

• SAFECHAIN™ Remediation Verification Test™;

• SAFECHAIN™ Completion-versus-Correction Principle™;

• SAFECHAIN™ Evidence-Led Reclassification Principle™;

• SAFECHAIN™ Immediate Escalation Principle™;

• SAFECHAIN™ Cross-Institution Accountability Assessment™;

• SAFECHAIN™ Human Accountability Continuity Test™;

• SAFECHAIN™ Assessment Gaming Principle™;

• SAFECHAIN™ Positive Accountability Evidence™;

• SAFECHAIN™ Neutral Assessment Principle™;

• SAFECHAIN™ Assessment Quality Review™;

• SAFECHAIN™ Accountability Integrity Assessment Record™;

• SAFECHAIN™ AIM-001 Accountability Integrity Dashboard™;

• SAFECHAIN™ Four-Layer Assessment Model™;

• SAFECHAIN™ AIM-001 Assessment Integrity Test™;

• and associated governance assessment, accountability, safeguarding, audit, assurance, monitoring, certification, accreditation, training and implementation materials.

No part of this publication may be reproduced, copied, republished, adapted, translated, distributed, licensed, sublicensed, sold, commercially exploited or incorporated into another governance assessment framework, accountability classification methodology, institutional-rating system, safeguarding architecture, audit programme, assurance methodology, certification scheme, accreditation programme, consultancy methodology, training product, artificial-intelligence system, analytics platform, software product, digital platform or derivative commercial offering without prior written permission from the applicable rights holder, except to the extent otherwise permitted by applicable law.

Publication or public accessibility of AIM-001™ does not grant permission to conduct official SAFECHAIN™ Accountability Integrity Assessments™, award AI1™–AI5™ classifications, issue SAFECHAIN™ ratings, certificates, seals or credentials, or represent any person, organisation, service, assessor, consultancy, software platform or methodology as SAFECHAIN™ authorised, certified or accredited.

No unauthorised person, organisation, consultant, auditor, investigator, assessor, certification body, accreditation body, training provider, technology provider or other entity may represent itself as authorised to conduct formal AIM-001™ assessments or award SAFECHAIN™ Accountability Integrity Classifications™ unless such authority has been expressly and validly granted under applicable SAFECHAIN™ governance, certification, accreditation and licensing arrangements.

References within AIM-001™ to generally established concepts including governance, accountability, evidence, materiality, safeguarding, audit, assurance, sampling, investigations, conflicts of interest, remediation, verification, oversight and organisational learning do not constitute claims of exclusive ownership over those underlying concepts.

The proprietary claim relates to the original SAFECHAIN™ expression, selection, arrangement, architecture, terminology, methodologies, classifications, tests and framework materials developed by the author.

The use of the ™ symbol identifies names, concepts, methodologies and framework identifiers being asserted as proprietary brand or framework designations. It does not, by itself, constitute a representation that any particular designation has been registered as a trade mark in any jurisdiction.

Nothing within AIM-001™ should be interpreted as legal advice, statutory guidance, regulatory approval, governmental accreditation, a determination of legal liability or a substitute for applicable legislation, regulation, professional standards, safeguarding duties, contractual requirements or binding governance instruments.

Any AI1™–AI5™ classification produced through AIM-001™ should be represented only within the precise scope, assessment period, evidence base, methodology, confidence level, organisational context, assumptions, exclusions and limitations actually assessed.

An AIM-001™ assessment or AI classification does not, by itself, establish negligence, professional misconduct, criminal responsibility, regulatory breach, discrimination, bad faith, breach of statutory duty or other legal liability.

Similarly, an AI1™ classification should not be represented as a guarantee that future accountability failure cannot occur.

Classification represents the evidence-based accountability condition demonstrated within the assessed scope and period.

Author and Framework Developer:
Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA
Founder — SAFECHAIN™

Framework: The SAFECHAIN™ Accountability Integrity Assessment Methodology™
Framework Reference: AIM-001™
Parent Framework: ACCOUNTABILITY-001™
Classification Architecture: AI1™–AI5™
Framework Series: SAFECHAIN™ Accountability Integrity Series
Version: 1.0
Year: 2026
Copyright: © 2026 Samantha Avril-Andreassen. All Rights Reserved.