Auditable Trustworthiness LevelsHTML teaching deck · 26 slides
01Research teaching deck

Auditable Trustworthiness Levels in AI Lifecycle Governance

A practical method for turning high-level AI trustworthiness commitments into monitored, documented, and reviewable lifecycle states.

Teaching notes
  • Source paper: Andrea Ferrario, arXiv:2607.16130v1, 17 Jul 2026.
  • Frame: not a replacement for legal, domain, risk, or safety judgment; an evidential layer for governance decisions.
Protocol
Profile
Level
Drift
Q(t) → Level
02Core thesis

Trustworthiness is not a static score. It is a governed lifecycle state.

The paper’s central move is to translate socio-technical trustworthiness into measurable profiles, map those profiles to auditable levels, then monitor level changes, margins, and drift over time.

Teaching notes
  • The methodology sits between raw MLOps metrics and governance decisions.
  • It helps answer: did the system remain within its documented trustworthiness regime after updates, shocks, or context shifts?
Design-time standardProtocol + labels
Monitoring layerProfiles + rule
Governance actionReview + reassessment
03Why this matters

The governance gap: principles are broad; metrics are narrow.

Existing frameworks often identify requirements like robustness, fairness, privacy, transparency, safety, and accountability. Operational systems often track local metrics. The missing layer is an auditable rule that connects heterogeneous measurements to governance-relevant trustworthiness levels over time.

Teaching notes
  • This is the “translation layer” claim.
  • It is especially relevant when conformity assessment, post-deployment monitoring, and reassessment obligations exist.
Principlesfairness
robustness
transparency
missing auditable translation layer
Operationslogs
metrics
incidents
04Paper contributions

Four concrete contributions

The paper contributes a formal framework, a lifecycle governance procedure, two diagnostics, and a proof-of-concept simulation study.

Teaching notes
  • The paper does not claim external validation.
  • The simulations demonstrate behavior under controlled lifecycle events: degradation, shocks, updates, asynchronous cadences, and system-copy comparison.
1Formal framework
2Lifecycle procedure
3Margins + drift
4Simulation proof of concept
05Concept map

The vocabulary you need

The methodology is easiest to learn as a pipeline: context defines the protocol; measurements form profiles; labels define levels; a rule assigns levels; diagnostics show fragility and movement.

Teaching notes
  • Protocol = selected trustworthiness dimensions for an intended purpose and context.
  • Profile = observed dimension scores at time t.
  • Level = governance-relevant class such as risky, acceptable, good, excellent.
ContextProtocolProfileRuleLevelAction
06Protocol

A trustworthiness protocol is not universal. It is context-bound.

For each AI system, governance actors select dimensions that matter for the intended purpose and context of use. A recommender, clinical tool, lending model, and agentic compliance system should not share an identical trustworthiness protocol by default.

Teaching notes
  • The paper treats trustworthiness as objective relative to a documented governance standard, not as a universal property.
  • The protocol must record inclusion rationale, omitted dimensions, cadence, limitations, and evidence windows.
Intended purpose
Context of use
Risk profile
Domain duties
Evidence sources
Review cadence
07Profile

Measurements become a normalized trustworthiness profile.

Each dimension can have one or more sub-metrics. Sub-metrics are normalized to a bounded scale, aggregated into dimension scores, then collected as Q(t), the current profile in trustworthiness space.

Teaching notes
  • Higher is not always better. Fairness gaps are often lower-is-better.
  • Some measures come from logs immediately. Others require delayed outcomes, audits, or review windows.

Protocol: PA = {d1, …, dn}

Dimension score: Qi(t) = ai(qi,1, …, qi,k)

Profile: QA(t) = (Q1(t), …, Qn(t)) ∈ [0,1]n

08Dimensions

Different dimensions have different evidence clocks.

Trustworthiness monitoring is usually discrete and asynchronous. Accuracy may depend on resolved outcomes. Explainability may be periodic. Robustness may combine drift indicators and red-team tests. Fairness may require batch review and protected-attribute governance.

Teaching notes
  • This is a major design implication: “latest available” evidence must be versioned and timestamped per dimension.
  • The artifact must make cadence and delay visible.
Continuous-ishlogs, override rate, latency
Windowedaccuracy, calibration, fairness
Periodicred-team, privacy audit, sustainability
Delayedoutcomes, appeals, incident resolution
09Empirical rule

The learned rule turns expert judgment into a reusable monitoring rule.

Experts define anchor regions and labels during design-time review. A simple interpretable learner, such as a shallow decision tree, can then learn threshold-style rules that assign new profiles to trustworthiness levels.

Teaching notes
  • The learner extends costly expert adjudication to routine monitoring.
  • The rule may preserve, compress, refine, or fail to improve the expert standard.
Expert anchor regions
good / acceptable / risky
Labeled dataset
{Q(ti), Yi}
Empirical rule
T̂(Q) → level
10Decision trees

Why decision trees are a reasonable proof-of-concept

Many governance decisions are naturally expressed as thresholds: performance below alpha, robustness above beta, fairness gap below delta. Shallow trees produce explicit level plateaus and readable transition boundaries.

Teaching notes
  • Decision trees are a choice, not a requirement.
  • Use more expressive models when rules are high-dimensional, interaction-heavy, or non-axis-aligned.
qacc ≥ 0.60?
Level 1
risky
qrob ≥ 0.60?
Level 2
acceptable
Level 3
good
11Labeling

The hard part is not the classifier. It is stable reference labels.

The paper assumes labels come from documented expert adjudication. This is costly, sparse, delayed, and institution-specific. Borderline profiles may remain unassigned until further review.

Teaching notes
  • Label protocol versioning is central.
  • The audit dataset should record who labeled what, under which protocol, with what disagreement process.
Costlyexpert review
Sparsefew regimes covered
Delayedoutcomes arrive later
Versionedstandards evolve
12Lifecycle procedure

The lifecycle workflow has three phases.

Phase 1 defines the governance standard. Phase 2 learns or rejects a candidate rule. Phase 3 monitors profiles and triggers review, validation, dataset updates, or reassessment.

Teaching notes
  • Post-deployment predictions are not automatically new labels.
  • Dataset update requires documented validation and relabeling under a protocol.
Phase 1Define protocol, measures, labels
Phase 2Learn or retain operative rule
Phase 3Monitor, report, reassess
13Controls

Human decision rights remain explicit.

The methodology assigns evidence production, technical implementation, independent challenge, governance approval, monitoring, and assurance. It uses familiar controls: segregation of duties, escalation rules, approval gates, audit records, and corrective actions.

Teaching notes
  • The method does not require an exotic governance body.
  • It does require clear ownership and evidence retention.
Owner
Domain experts
Technical team
Independent review
Governance authority
Assurance
14Diagnostic 1

Boundary margin measures local fragility.

Boundary margin asks: how close is the current profile to leaving its current trustworthiness plateau? A large margin means the system is well inside its assigned level. A small margin means small measurement variation or degradation could trigger a level transition.

Teaching notes
  • This is spatial closeness to a learned regime boundary.
  • The distance function and alert thresholds are governance choices, not universal constants.
Level 2
Level 3
small margin
large margin
15Diagnostic 2

Profile drift measures movement over time.

Profile drift asks: how far has the trustworthiness profile moved over a chosen time horizon? Drift can spike after shocks, updates, environmental changes, or new measurement windows.

Teaching notes
  • Margin and drift are complementary.
  • A system can be stable but near a boundary, or far from a boundary but moving quickly.
shock
update
D(t;h) = distance between Q(t) and Q(t-h)
16Reporting artifact

The reporting artifact is the system of record.

The paper proposes a compact artifact that records context, protocol, measures, label scale, learned rule, validation, data adequacy, monitoring state, responsibility, evidence, and reassessment conditions.

Teaching notes
  • This artifact bridges conformity documentation, post-deployment monitoring, and reassessment.
  • In enterprise use, this maps naturally to an AI control plane, evidence ledger, model card extension, or governance registry.
System context
Protocol
Measures
Labels
Rule + validation
Current state
Escalation
Evidence record
17Experiment 1

Dense recommender vs. sparse clinical system

With the same two-dimensional profile structure, dense monitoring produced near-perfect rule recovery. Sparse clinical monitoring produced a more jagged trajectory and compressed trustworthiness regimes under the selected simple tree.

Teaching notes
  • Recommender: 720 dense observations, 3 levels.
  • Clinical: 48 sparse observations, 4 levels.
  • Finding: cadence and regime coverage materially affect what can be learned.
Recommender720 dense observationsrule recovered cleanly
Clinical48 sparse observationsjagged, coarser regimes
18Experiment 2

Asynchronous lending metrics remain interpretable, but class imbalance matters.

A loan-creditworthiness system used four dimensions with different cadences: performance, explainability, robustness, and fairness. The learned tree was compact and mostly tracked high trustworthiness, but rare lower regimes were weakly supported.

Teaching notes
  • 90 governance-review profiles; most were level 4.
  • Test accuracy remained reasonable, but balanced accuracy and macro-F1 were weaker because lower levels were rare.
Performance every 4
Explainability every 12
Robustness every 12
Fairness every 16
Latest-available profile at review time
19Experiment 3

Same AI system, different deployment context, different governance value.

Hospital Alpha stayed in the highest regime under a coarse expert rule; learning added little. Hospital Beta moved across regimes under shocks and updates; learning was useful because it exposed compression, boundary fragility, and post-deployment instability.

Teaching notes
  • Alpha: stable environment, expert rule remains operative.
  • Beta: brittle environment, four reference levels but learned tree predicted only levels 2, 3, and 4.
Hospital Alphastable contextexpert rule retained
Hospital Betabrittle contextlearning exposes fragility
20Interpreting results

The method is valuable even when learning is imperfect.

A failed, compressed, or fragile learned rule is not just a modeling problem. It is governance evidence: the organization may lack regime coverage, label stability, measurement adequacy, or protocol maturity.

Teaching notes
  • The model is diagnostic of the governance substrate.
  • A constant classifier can be the correct conclusion when there is no informative lifecycle variation.
Compressionscale may be too fine for data
Boundary fragilitystates near transitions
Instabilitytest degradation under lifecycle shift
21Red-team view

What the methodology does not solve

It does not replace expert judgment. It does not automatically prove compliance. It does not solve weak metrics, noisy labels, missing protected attributes, protocol drift, class imbalance, or unstable deployment context.

Teaching notes
  • The paper calls the method lightweight by design.
  • Future work: real domain cases, other interpretable models, label-noise robustness, protocol drift, multi-agent workflows.
Weak metrics
Noisy labels
Class imbalance
Protocol drift
Missing evidence
Overfit boundaries
22Architecture implication

Think of this as an AI governance instrument panel.

Raw metrics are sensors. The protocol is the calibration standard. The learned rule is the gauge. Boundary margin is proximity to a threshold. Profile drift is movement over time. The governance artifact is the flight recorder.

Teaching notes
  • This metaphor helps executives and engineers align.
  • The system is only useful if the sensors, calibration, and response process are credible.
Sensorsmetrics
Gaugerule
Altimeterlevel
Flight recorderevidence artifact
23Implementation blueprint

Minimal enterprise implementation pattern

Start with a registry-backed protocol, versioned measurements, expert-labeled anchor cases, an interpretable rule service, diagnostics, alert thresholds, review queues, and an evidence ledger.

Teaching notes
  • Use existing observability and model-risk controls where possible.
  • Keep the rule auditable before making it more expressive.
Protocol registry
Metric pipelines
Label workbench
Rule service
Diagnostics
Review queue
Evidence ledger
24Controls to add

Make the governance loop operational, not decorative.

Every level transition should have a defined response: no action, intensified monitoring, expert review, relabeling, rollback, restricted deployment, documentation update, or reassessment.

Teaching notes
  • The response catalog is where governance becomes operational.
  • This is also where accountability and human authorization must be explicit.
No actionIntensify monitoringExpert reviewRelabelRetrainRollbackRestrictReassess
25Key takeaways

What to remember

The paper provides a practical bridge between trustworthy-AI principles and lifecycle governance. Its biggest value is not the decision tree. It is the disciplined translation from context to protocol to profile to level to diagnostic to auditable action.

Teaching notes
  • Best next step: pilot this method on one AI product with mature telemetry and clear governance owners.
  • Avoid using it as a compliance shortcut. Use it as evidence infrastructure.
1Define trustworthiness by context.
2Make level rules explicit.
3Monitor margins and drift.
4Escalate with evidence.
26Provenance

Sources and grounding

This deck is a teaching synthesis of the paper plus minimal context from official governance sources. It separates the paper’s claims from implementation interpretation and enterprise application guidance.

Teaching notes
  • Primary source: Ferrario, “A Methodology for Auditable Trustworthiness Levels in AI Lifecycle Governance,” arXiv:2607.16130v1.
  • Context: EU AI Act Regulation 2024/1689 and NIST AI RMF 1.0 / Playbook.

Primary paper
Ferrario, A. (2026). A Methodology for Auditable Trustworthiness Levels in AI Lifecycle Governance. arXiv:2607.16130v1.

Context sources
EU AI Act Regulation (EU) 2024/1689; NIST AI Risk Management Framework 1.0 and NIST AI RMF Playbook.

Deck status
Generated 2026-07-20 18:17 UTC. Teaching synthesis; not legal advice or compliance certification.

← / → to navigate · Space next · T theme