AI EngineeringZero to ProductionHome·About·Contact
Enterprise & Leadership · Part 1

AI governance & compliance

You already ship AI. This lesson is the layer above the code: the governance program that lets an organization deploy AI systems it can defend to a regulator, an auditor, and its own board. We map genuinely public frameworks — the EU AI Act risk tiers, ISO/IEC 42001, and the NIST AI Risk Management Framework — onto engineering artifacts you can actually produce: model cards, audit trails, and a RACI that says who owns what. Framework-as-fact, attributed; company examples are representative scenarios. This is engineering-leadership framing, not legal advice.

⏱️ ~90 min🏢 Org / decision layer🎯 Lead / staff

Learning objectives

  • Explain what an AI governance program is and why it is an engineering concern, not just a legal one.
  • Place a system into the EU AI Act risk tiers (prohibited / high-risk / limited / minimal) and name the duties each tier implies — verifying specifics against the current text.
  • Map controls to ISO/IEC 42001 (an AI management-system standard) and the NIST AI RMF functions (Govern, Map, Measure, Manage).
  • Produce the two artifacts auditors actually ask for: a model card and an audit trail.
  • Assign accountability with a RACI so 'who signs off on this model' has one answer.
What this lesson is — and isn'tWe cite public frameworks (EU AI Act, ISO/IEC 42001, NIST AI RMF, SOC 2, GDPR) as fact and attribute them. Regulatory specifics drift: treat every tier boundary, date, and obligation here as "verify against the current regulation/text." Company situations are labelled representative scenarios — not real named-customer stories. Nothing here is legal advice; confirm obligations with your legal/compliance function.

1 · What a governance program actually is

A governance program is the standing machinery that decides which AI systems may exist, under what controls, and who is accountable when one misbehaves. For an engineer it is not paperwork for its own sake — it is the difference between "we think this is fine" and "we can show a regulator exactly why this is fine." The three moving parts are policy (what we will and won't build), process (the gates a system passes before and after launch), and evidence (the artifacts that prove the process ran).

The failure mode governance prevents is the orphaned model: a system in production that no one owns, whose training data no one can describe, and whose behavior no one is monitoring. Every framework below is, at heart, a structured way to make orphaned models impossible.

Governance is an interface, not a gate you resentGood governance is designed like an API: clear inputs (a model card, an eval report), clear outputs (approve / approve-with-conditions / reject), and a fast path for low-risk systems so the process scales. If your governance blocks a spellchecker as hard as a credit-decision model, engineers will route around it — and then you have no governance at all.

2 · EU AI Act — risk tiers as a design input

The EU AI Act is public, adopted law that classifies AI systems by risk tier and attaches escalating obligations. As an engineer, treat the tier as a design input the way you treat a latency budget: it changes what you must build. The four commonly-described tiers:

TierRough meaningIllustrative obligation (verify current text)
ProhibitedUnacceptable-risk uses (e.g. certain social scoring, some biometric practices).Not allowed to place on the market — a hard no, not a control.
High-riskSystems in sensitive domains (e.g. specified employment, credit, critical-infrastructure uses).Risk management, data governance, logging, human oversight, technical documentation, conformity assessment.
Limited-riskSystems that interact with people or generate content.Transparency: tell users they're dealing with AI / that content is AI-generated.
Minimal-riskEverything else (e.g. spam filters, most game AI).No specific obligations beyond existing law.
Tier boundaries and dates drift — verifyThe exact categories, the list of high-risk use cases, and the phased application dates have specific legal definitions and timelines that evolve. Do not memorize a boundary from this page; classify against the current published Act and its annexes, with your legal team. The durable engineering lesson is the shape: higher risk → more documentation, more human oversight, more logging.

Note the asymmetry engineers miss: an EU-facing product may be high-risk because of where it is used, not because of how clever the model is. A boring logistic-regression that decides loan eligibility can carry heavier obligations than a frontier LLM writing marketing copy.

3 · ISO/IEC 42001 & NIST AI RMF — the control catalogs

Two public frameworks tell you how to run governance rather than what is illegal. ISO/IEC 42001 is an international standard for an AI management system (AIMS) — the same management-system pattern as ISO 27001 for security, applied to AI: define policy, assign roles, run a plan-do-check-act cycle, and keep records. It is certifiable, which matters for enterprise sales. The NIST AI RMF is a voluntary US framework organized around four functions you can use as a checklist:

NIST AI RMF functionThe question it forcesEngineering artifact
GovernWho is accountable and what is our risk appetite?Policy doc, RACI, model inventory
MapWhat is this system, its context, and its risks?Model card, intended-use statement, risk register entry
MeasureHow do we quantify performance, bias, and robustness?Eval suite, bias metrics, red-team results
ManageHow do we act on risk over the lifecycle?Monitoring, incident process, retirement plan
You don't pick one — you map themThese frameworks overlap on purpose. A single control — "every production model has an owner and a monitoring dashboard" — satisfies NIST Manage, an ISO 42001 clause, and part of an EU high-risk logging duty at once. Governance maturity is largely the work of writing one control set and mapping it to every framework you're measured against, so you audit once and report many times.

4 · The artifacts auditors ask for: model cards & audit trails

Frameworks become real as two artifacts. A model card is a short, standardized document describing a model's intended use, training-data provenance, evaluation results, known limitations, and owner. An audit trail is the immutable record of what happened: which model version served which request, what it was asked, what it returned (privacy-preserving), and who approved its deployment. Together they answer the auditor's two questions — "what is this system?" and "can you prove it behaved as claimed?"

A model card is a diff, not an essayThe most maintainable model card is generated from your pipeline: version, eval scores, and data hashes are filled in by CI, and only the intended-use and limitations prose is hand-written. When the model changes, the card changes automatically — an out-of-date model card is worse than none, because it is evidence that your process lies.

5 · Who owns the model? — the RACI

The single most common governance gap is diffuse ownership. A RACI matrix (Responsible, Accountable, Consulted, Informed) fixes this by naming, for each decision, exactly one Accountable role. The pipeline below shows the flow a system passes through and who owns each gate.

Intake register it Risk-tier classify Build + card evidence Review gate one approver Deploy ship Monitor & audit
DecisionResponsibleAccountableConsultedInformed
Risk-tier classificationML engineerProduct ownerLegal / complianceSecurity
Model-card sign-offML engineerEng leadData governanceProduct
Deploy approvalEng leadGovernance board / risk ownerLegal, SecurityExecs
Incident responseOn-call engEng leadLegal, CommsGovernance board
Exactly one 'A' per rowThe classic mistake is two Accountable owners — which means zero, because each assumes the other signed off. If a row can't name a single accountable role, the decision has no owner and the system is an orphaned model waiting to happen.

✓ Checkpoint — you can move on when you can…

  • Explain the difference between policy, process, and evidence in a governance program.
  • Given a system description, place it in an EU AI Act tier and name one obligation that tier implies (noting you'd verify the current text).
  • Map a single control to both a NIST AI RMF function and an ISO/IEC 42001 concern.
  • Say what a model card and an audit trail each prove to an auditor.
  • Fill a RACI row with exactly one Accountable owner and justify it.
✓ Knowledge check

A team wants to ship an internal tool that ranks job applicants' résumés for recruiters. An engineer says "it's just a classifier, low risk." Why is that reasoning dangerous under an EU AI Act lens, and what changes about how you'd build it?

Show answer
The risk tier is driven by the use, not the model class. Employment-related decision support is commonly cited as a high-risk use, which — verifying against the current text — typically pulls in duties like documented risk management, data governance, logging, human oversight, and technical documentation. So "just a classifier" is exactly backwards: a simple model in a high-stakes domain can carry heavier obligations than a complex one in a trivial domain. Concretely you'd add a model card, bias/robustness evals, an audit trail of each ranking, and a human-in-the-loop so a recruiter — not the tool — makes the decision. Confirm the actual classification with legal.
✓ Knowledge check

Your company is already SOC 2 compliant. Leadership asks, "doesn't that cover our AI governance?" Give a precise answer.

Show answer
No. SOC 2 attests to controls around security, availability, and confidentiality of a service — it says nothing about whether a model is fit for purpose, unbiased, well-documented, or used within legal limits. AI governance frameworks (ISO/IEC 42001, NIST AI RMF, the EU AI Act) address AI-specific risks: data provenance, evaluation, human oversight, model documentation, and lifecycle management. There is overlap — audit trails and access controls serve both — but SOC 2 is necessary-not-sufficient. The honest framing to leadership: SOC 2 proves we run the service responsibly; an AIMS proves we run the model responsibly.

🪜 Practice ladder beginner → industry

Six graded exercises, easy to real-world. Try each before opening its solution.

Exercise 1 · Classify three systems into EU AI Act tiersBeginner

Context: Risk-tiering is the first governance reflex: before you design controls you decide how much control is warranted.

Your task: Given three representative systems, assign each a plausible EU AI Act tier and name one obligation implied — flagging that you would verify against the current text.

Requirements:

  • System A: a spam filter for internal email
  • System B: an LLM chatbot on a public website that answers product questions
  • System C: a model that scores loan applications for approval
  • For each, state the tier and one duty, and note the verify-against-current-text caveat

💡 Hint: The tier is about the use and its stakes, not the model's sophistication.

Show solution

System A — spam filter → minimal-risk. No AI-Act-specific obligations beyond existing law; ordinary security/privacy hygiene applies.

System B — public product chatbot → limited-risk. The salient duty is transparency: users should be told they are interacting with an AI system. Verify the exact transparency wording against the current Act.

System C — loan-approval scoring → high-risk. Credit/eligibility decisions are a commonly-cited high-risk use, pulling in duties such as risk management, data governance, record-keeping/logging, human oversight, and technical documentation, plus a conformity assessment. Confirm the current annex and obligations with legal.

The lesson: the simplest model (C could be logistic regression) carries the heaviest obligations, because tiering follows stakes, not complexity.

Exercise 2 · Draft a model card for an existing systemIntermediate

Context: A model card is the single artifact every framework asks for and most teams don't have.

Your task: Write a model-card template with the sections an auditor expects, then fill it for one representative system.

Requirements:

  • Include: intended use, out-of-scope uses, training/reference data provenance, evaluation results, known limitations, and a named owner
  • Mark which fields should be auto-filled by CI vs hand-written
  • Keep it to one screen — a card no one maintains is worthless

💡 Hint: Separate the machine-generated facts (version, scores) from the human judgment (intended use).

Show solution

Model card template (representative fill for a support-triage classifier):

  • Model: support-triage-classifier v2.3.1 (auto: version, git SHA)
  • Owner: Support Platform team, accountable: Eng Lead (name/role) (hand-written)
  • Intended use: route inbound support tickets to one of 6 queues to assist human agents. (hand-written)
  • Out-of-scope: auto-closing tickets, any customer-facing action, non-English at launch. (hand-written — the most important section)
  • Training data: 120k historical tickets, 2022–2024, PII-stripped; provenance hash sha256:… (auto)
  • Evaluation: macro-F1 0.88 on held-out set; per-queue precision table; fairness check across language groups (auto from eval CI)
  • Limitations: degrades on tickets >2k tokens; unknown behavior on new product lines. (hand-written)

Why the split matters: the auto fields update on every retrain so the card can't lie; the hand-written fields are the human accountability that a framework actually cares about.

Exercise 3 · Map one control set to NIST AI RMF and ISO 42001Advanced

Context: Governance maturity is auditing once and reporting many times.

Your task: Take a set of five concrete engineering controls and map each to a NIST AI RMF function and an ISO/IEC 42001 concern, showing the overlap.

Requirements:

  • Controls: model inventory, eval-gated deploy, monitoring dashboard, incident runbook, data-provenance log
  • For each, name the NIST function (Govern/Map/Measure/Manage) it primarily serves
  • Note where one control satisfies multiple frameworks at once

💡 Hint: Most controls map to Measure or Manage; the org-level ones map to Govern.

Show solution

Crosswalk (verify clause references against the current standards):

ControlNIST AI RMFISO/IEC 42001 concern
Model inventory (every prod model registered + owned)GovernAIMS scope, roles & responsibilities, asset management
Eval-gated deploy (CI fails deploy on metric regression)Measure + ManagePerformance evaluation, operational controls
Monitoring dashboard (drift, error rate, refusal rate)ManageMonitoring, measurement & continual improvement
Incident runbook + post-mortem-to-eval loopManage + GovernNonconformity & corrective action
Data-provenance log (source, license, PII status)MapData governance, documented information

Takeaway: five controls, three frameworks, one implementation. The eval-gated deploy alone touches NIST Measure and Manage and two ISO clauses — build it once, report it three ways.

Exercise 4 · Design the review-gate process (fast path + full path)Expert

Context: Governance that treats a spellchecker like a credit model gets bypassed.

Your task: Design a two-lane review gate: a fast path for low-risk systems and a full path for high-risk ones, with explicit entry criteria and required evidence for each.

Requirements:

  • Define the trigger that sends a system down each lane
  • List the evidence required at each gate
  • Name the accountable approver per lane
  • Include an escalation rule if a fast-path system is later found to be higher-risk

💡 Hint: The lane is chosen by risk tier + blast radius (reversible? touches money/health/people?).

Show solution

Lane selection. Score two axes at intake: risk tier (EU-Act-style) and blast radius (can it take an irreversible or customer-facing action?). Low tier + low blast radius → fast path; anything high on either axis → full path.

Fast path (e.g. internal spam filter). Evidence: a one-page model card + a link to the eval run. Approver: Eng Lead. SLA: async, 1 business day. Auto-approve if eval gates pass and no PII/PHI is touched.

Full path (e.g. loan scoring). Evidence: full model card, bias/robustness eval report, red-team results, data-provenance log, human-oversight design, monitoring plan, rollback plan. Approver: Governance board / risk owner, with Legal and Security consulted. SLA: scheduled review.

Escalation rule. If monitoring, an incident, or a use-case change reveals a fast-path system now has higher blast radius (e.g. someone wired the spam filter to auto-delete), it is immediately reclassified to the full path, its deploy is frozen pending review, and the misclassification becomes a governance post-mortem. The fast path is a privilege contingent on staying low-risk.

Exercise 5 · Write the one-page AI governance policyProfessional

Context: A policy no one can read in five minutes is a policy no one follows.

Your task: Draft a one-page AI governance policy for a mid-size company that an engineer, a lawyer, and an exec can all act on.

Requirements:

  • State scope: what counts as an 'AI system' under the policy
  • State the risk-tiering rule and the review lanes
  • State the mandatory artifacts (model card, audit trail, owner)
  • State the frameworks you align to and the verify-against-current-text stance
  • Keep it to one page; link out to detailed procedures

💡 Hint: Policy = the durable rules; procedures = the changeable how-to. Don't mix them.

Show solution

AI Governance Policy (v1) — representative one-pager.

1. Scope. Any system that uses machine learning or an LLM to inform a decision, generate content, or take an action on behalf of the company — built or bought. Excludes purely deterministic rules.

2. Risk tiering. Every AI system is classified at intake into minimal / limited / high risk using the current EU AI Act tiers as a reference, plus a blast-radius check. Classification is recorded in the model inventory.

3. Review lanes. Low-risk systems take the fast path (Eng Lead approval + model card). High-risk or high-blast-radius systems take the full path (Governance board approval + full evidence pack).

4. Mandatory artifacts. No system reaches production without (a) a named accountable owner, (b) a current model card, and (c) an audit trail. These are deploy blockers.

5. Framework alignment. We align our control set to ISO/IEC 42001 and the NIST AI RMF, and we classify EU-facing systems under the EU AI Act. Regulatory specifics are verified against the current text with Legal; this policy is not legal advice.

6. Review cadence. Policy reviewed twice a year and after any material regulatory change. Detailed procedures (templates, gates, RACI) live in linked runbooks so the policy itself stays stable.

Exercise 6 · Stand up a governance review for a real proposed systemIndustry scenario

Context: Representative scenario: a mid-size fintech proposes an LLM feature that drafts explanations of loan decisions for customers. You chair the review.

Your task: Run the full governance review end to end and produce the artifacts and decision an auditor would expect to see.

Requirements:

  • Classify the system and justify the tier (verify against current text)
  • List the evidence you require before approval
  • Identify the top three risks and the control mitigating each
  • Fill the RACI for build, deploy, and incident response
  • Issue a decision: approve / approve-with-conditions / reject, with reasons
  • Name what monitoring must exist post-launch

💡 Hint: The feature 'only explains' a decision — but an inaccurate explanation of an adverse credit decision is itself a regulated, high-stakes harm.

Show solution

Classification. Although the model "only explains," it is tightly coupled to a credit decision and its output goes to consumers about an adverse action — treat it as high-risk (verify against the current Act with Legal). A wrong or misleading explanation can mislead a consumer and create legal exposure under credit/consumer-protection law.

Required evidence. Full model card; faithfulness eval proving the explanation matches the actual decision factors (not a plausible confabulation); bias check across protected groups; red-team for prompt injection and for explanations that contradict the decision; data-provenance log; human-oversight design; rollback plan.

Top three risks + controls. (1) Confabulated rationale — explanation doesn't reflect the real model → force the explanation to be generated from the decision model's actual feature attributions, and gate on a faithfulness eval. (2) Inconsistent/biased explanations across groups → bias eval + monitoring on explanation content by segment. (3) Scope creep — feature starts advising customers how to 'fix' their application → block advice-language, decision-support framing only, in code.

RACI. Build: Responsible = ML eng, Accountable = Eng Lead, Consulted = Legal/Compliance, Informed = Product. Deploy: Accountable = Governance board / risk owner. Incident: Accountable = Eng Lead, Consulted = Legal + Comms.

Decision: Approve-with-conditions — ship only after the faithfulness and bias evals pass a preset bar, with a human able to review any explanation, and with the advice-language block in place. Reject if faithfulness can't be demonstrated: an explanation you can't prove is worse than none.

Post-launch monitoring: faithfulness sampling, explanation-content drift, complaint rate, and segment-level fairness — feeding the risk register and a scheduled re-review.

© 2026 studybydoing.in · AI Engineering: Zero to Production · All rights reserved. · About · Privacy Policy · Terms · Contact
Educational content, provided as-is and without warranty. Code samples are examples — review, test, and adapt them before using in production. See the Terms of Use & Disclaimer. Use at your own risk.
© studybydoing.in