Skip to main content
CertRampLearning

PECB · Free practice questions

ISO/IEC 42001 Foundation Practice Questions

15 original ISO/IEC 42001 Foundation practice questions across both PECB competency domains, weighted like the real exam: 6 on fundamental principles and concepts of an AIMS and 9 on the AI management system requirements. Each one comes with the answer, a short explanation and the reason every other option is wrong. The real exam uses three options per question; these use four to make you work a little harder.

Written for the current ISO/IEC 42001 Foundation exam · Updated November 11, 2026

These are original practice questions written by CertRamp to the official PECB exam outline. They are not real exam questions and do not come from any exam. Try each question first, then open the answer to see the explanation for every option.

Domain 1

Fundamental principles and concepts of an artificial intelligence management system (AIMS)

Purpose, scope and applicability of ISO/IEC 42001FoundationDomain: Fundamental principles and concepts of an artificial intelligence management system (AIMS)

Question 1. A 40-person logistics company does not build AI, but it uses a purchased AI route-planning service in daily operations. Can it implement an AIMS based on ISO/IEC 42001?

  1. Option A: Yes. It applies to any organization that provides or uses AI
  2. Option B: No. Only organizations that develop and train their own AI models are in scope
  3. Option C: Only if it already holds an ISO/IEC 27001 certificate for its information security management system
  4. Option D: No. The standard is intended for large enterprises and public authorities
Show answer and explanation

Correct answer: A. Yes. It applies to any organization that provides or uses AI

ISO/IEC 42001 is a generic management system standard. Developers, providers and users of AI can all establish an AIMS, scaled to their size and the AI systems in scope.

  • A (correct)Correct. ISO states that ISO/IEC 42001 applies to organizations of any size involved in developing, providing or using AI-based products or services.
  • B (incorrect)Using AI is explicitly covered, not only developing it.
  • C (incorrect)ISO/IEC 42001 can be implemented on its own; an existing ISO/IEC 27001 system helps integration but is not a precondition.
  • D (incorrect)ISO describes the standard as applicable regardless of organization size.

Reference: ISO/IEC 42001:2023 Information technology: Artificial intelligence: Management system (opens in a new tab)

Machine learning approaches, including supervised learning and deep learningFoundationDomain: Fundamental principles and concepts of an artificial intelligence management system (AIMS)

Question 3. A bank trains a model on 10 years of past loan applications, each labelled with whether the borrower later defaulted, so that it can predict default for new applications. Which machine learning approach is this?

  1. Option A: Unsupervised learning
  2. Option B: Supervised learning
  3. Option C: Reinforcement learning
  4. Option D: A rule-based expert system
Show answer and explanation

Correct answer: B. Supervised learning

Supervised learning uses labelled training data, which is typical for classification and regression tasks such as credit scoring. The quality and representativeness of those labels is also where many AI risks, such as bias, start.

  • A (incorrect)Unsupervised learning finds structure such as clusters in unlabelled data; here every example carries a known outcome.
  • B (correct)Correct. The model learns from input examples paired with known labels (defaulted or not) and predicts the label for new inputs.
  • C (incorrect)Reinforcement learning learns from rewards received while acting in an environment, not from a fixed labelled data set.
  • D (incorrect)An expert system applies rules written by people; it does not learn its behaviour from labelled data.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

Structure, layers and functions of artificial neural networksFoundationDomain: Fundamental principles and concepts of an artificial intelligence management system (AIMS)

Question 4. What best distinguishes deep learning from other machine learning approaches?

  1. Option A: It can only be used for image recognition
  2. Option B: It does not need any training data
  3. Option C: Its decision logic is written as explicit if-then rules by domain experts
  4. Option D: It uses neural networks with many hidden layers
Show answer and explanation

Correct answer: D. It uses neural networks with many hidden layers

A neural network has an input layer, one or more hidden layers and an output layer; deep learning uses many hidden layers. ISO/IEC 23053 treats deep learning as a subset of machine learning.

  • A (incorrect)Deep learning is also used for language, speech, tabular data and more.
  • B (incorrect)Deep learning typically needs large amounts of training data.
  • C (incorrect)That describes symbolic, rule-based AI; deep learning learns its parameters from data.
  • D (correct)Correct. Deep learning is a subset of machine learning based on neural networks with multiple hidden layers, which learn increasingly abstract features.

Reference: ISO/IEC 23053:2022 Framework for AI systems using machine learning (opens in a new tab)

Management system concepts and their fundamental elementsIntermediateDomain: Fundamental principles and concepts of an artificial intelligence management system (AIMS)

Question 5. ISO/IEC 42001 follows the Plan-Do-Check-Act approach. In which phase do internal audits and management reviews of the AIMS mainly belong?

  1. Option A: Plan
  2. Option B: Do
  3. Option C: Check
  4. Option D: Act
Show answer and explanation

Correct answer: C. Check

Mapping clauses to PDCA is a quick way to orient yourself: planning up front, clause 8 for doing, clause 9 for checking and clause 10 for acting. Management review outputs then feed decisions on improvement.

  • A (incorrect)Plan covers setting the AI policy, objectives, risk and impact assessments and planned actions (clauses 4 to 7 broadly).
  • B (incorrect)Do is operating the planned processes and controls (clause 8 Operation).
  • C (correct)Correct. Monitoring, measurement, internal audit and management review sit in clause 9 Performance evaluation, the Check phase.
  • D (incorrect)Act is acting on the results through corrective action and continual improvement (clause 10 Improvement).

Reference: ISO/IEC 42001:2023 Information technology: Artificial intelligence: Management system (opens in a new tab)

Principles of responsible and trustworthy AI, including transparencyFoundationDomain: Fundamental principles and concepts of an artificial intelligence management system (AIMS)

Question 6. An insurer tells customers when they are talking to an AI chatbot and gives them a plain-language explanation of the main factors behind automated claim decisions. Which responsible AI principle does this mainly support?

  1. Option A: Transparency
  2. Option B: Robustness
  3. Option C: Data minimization
  4. Option D: Environmental sustainability
Show answer and explanation

Correct answer: A. Transparency

Transparency means people can know that AI is involved and understand, at an appropriate level, how outcomes are produced. It also underpins accountability, because decisions that cannot be explained are hard to challenge.

  • A (correct)Correct. Disclosing that AI is used and explaining how it reaches decisions supports transparency and explainability towards affected people.
  • B (incorrect)Robustness is about the system performing reliably under varied or adverse conditions, not about informing users.
  • C (incorrect)Data minimization limits the personal data collected and processed; the measures described do not change the data used.
  • D (incorrect)Sustainability concerns the resource use of AI systems, which the described measures do not address.

Reference: ISO: AI management systems, what businesses need to know (opens in a new tab)

Domain 2

Artificial intelligence management system (AIMS)

Organizational context and interested-party requirements under ISO/IEC 42001FoundationDomain: Artificial intelligence management system (AIMS)

Question 7. An organization starting its AIMS determines the external and internal issues relevant to its use of AI, the interested parties and their requirements, and its role (for example AI provider or AI user). Which clause of ISO/IEC 42001 does this work address?

  1. Option A: Clause 6 Planning
  2. Option B: Clause 4 Context of the organization
  3. Option C: Clause 5 Leadership
  4. Option D: Clause 8 Operation
Show answer and explanation

Correct answer: B. Clause 4 Context of the organization

The context analysis drives everything else: the AIMS scope, the risks to address and the requirements of interested parties. In ISO/IEC 42001 it also includes the roles the organization plays in relation to its AI systems.

  • A (incorrect)Planning uses the results of the context analysis to address risks and set objectives, but determining the context itself comes earlier.
  • B (correct)Correct. Clause 4 covers understanding the organization and its context, the needs of interested parties and the scope of the AIMS.
  • C (incorrect)Clause 5 covers top management commitment, the AI policy and roles; it is not where the context analysis sits.
  • D (incorrect)Clause 8 covers operational planning and control, including carrying out assessments, once the system is set up.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

Top management's leadership and commitment responsibilitiesIntermediateDomain: Artificial intelligence management system (AIMS)

Question 8. Which of the following is a leadership responsibility of top management under ISO/IEC 42001?

  1. Option A: Personally carrying out every AI risk assessment and AI system impact assessment in the organization
  2. Option B: Acting as internal auditor for the AIMS processes they manage
  3. Option C: Approving each individual change to an AI model's code
  4. Option D: Ensuring the AI policy and AI objectives fit the strategic direction
Show answer and explanation

Correct answer: D. Ensuring the AI policy and AI objectives fit the strategic direction

Leadership in a management system means direction, resources and accountability rather than doing every task. Top management sets the AI policy, makes sure roles are assigned and supports continual improvement.

  • A (incorrect)Top management must ensure the AIMS works, but tasks such as risk assessments can be assigned to competent people.
  • B (incorrect)Auditors should be objective and impartial; auditing your own work conflicts with that.
  • C (incorrect)Change approval is an operational process that can be delegated; it is not a defined leadership duty.
  • D (correct)Correct. Clause 5 requires top management to show leadership, including establishing the AI policy and objectives and integrating the AIMS into business processes.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

Annex A control applicability, AI risk treatment and statement of applicability requirementsIntermediateDomain: Artificial intelligence management system (AIMS)

Question 9. After AI risk treatment, an organization concludes that some Annex A controls are not needed for its AI systems. What does ISO/IEC 42001 expect?

  1. Option A: Justify each inclusion and exclusion in the statement of applicability
  2. Option B: Implement every Annex A control anyway, because certification requires all of them to be in place
  3. Option C: Ask the certification body to remove the controls from the standard for that organization
  4. Option D: Leave the excluded controls out of all documentation so auditors do not ask about them
Show answer and explanation

Correct answer: A. Justify each inclusion and exclusion in the statement of applicability

Risk treatment decides which controls are needed, Annex A acts as a cross-check that nothing necessary was overlooked, and the statement of applicability documents the outcome. Organizations may also add controls from other sources.

  • A (correct)Correct. The statement of applicability lists the necessary controls and justifies inclusions and exclusions, so excluding a control is possible when it is justified.
  • B (incorrect)Annex A is a reference set of controls; the organization determines which are necessary based on risk treatment.
  • C (incorrect)Certification bodies audit against the standard; they do not change it per client.
  • D (incorrect)Exclusions must be documented and justified, not hidden.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

Conducting, documenting, reviewing and updating impact assessments, including consequences for individuals, groups and societyIntermediateDomain: Artificial intelligence management system (AIMS)

Question 10. What is the main focus of an AI system impact assessment under ISO/IEC 42001, compared with the AI risk assessment?

  1. Option A: It replaces the AI risk assessment for high-risk systems
  2. Option B: It estimates the financial return of an AI project
  3. Option C: Consequences of the AI system for individuals, groups and societies
  4. Option D: It measures the technical accuracy of the model on a held-out test data set before release
Show answer and explanation

Correct answer: C. Consequences of the AI system for individuals, groups and societies

The impact assessment widens the view beyond the organization to the people and societies affected by an AI system. It is documented, reviewed and updated when the system or its context changes.

  • A (incorrect)Both are required; the impact assessment does not replace the risk assessment.
  • B (incorrect)Business cases are useful but are not what the impact assessment requires.
  • C (correct)Correct. The impact assessment looks at how the AI system and its foreseeable uses can affect people and society, and its results feed into the risk assessment.
  • D (incorrect)Model testing is part of verification and validation in the life cycle, not the impact assessment.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

AI risk management process: establishing scope, context and criteria, then identifying, analysing, evaluating, prioritizing, treating and monitoring risksFoundationDomain: Artificial intelligence management system (AIMS)

Question 11. In the AI risk management process, which step compares the results of risk analysis with the organization's AI risk criteria to decide which risks need treatment and in what priority?

  1. Option A: Risk identification
  2. Option B: Risk evaluation
  3. Option C: Risk treatment
  4. Option D: Risk monitoring
Show answer and explanation

Correct answer: B. Risk evaluation

The sequence is identify, analyse, evaluate, then treat, with monitoring and review throughout. Defined AI risk criteria are what make evaluation consistent and repeatable.

  • A (incorrect)Identification finds and describes risks; it comes before they are analysed or compared with criteria.
  • B (correct)Correct. Risk evaluation compares analysed risk levels with the risk criteria to support decisions on treatment and priorities.
  • C (incorrect)Treatment selects and implements options to modify risk after evaluation has decided what needs action.
  • D (incorrect)Monitoring and review keep track of risks and controls over time; it is not the comparison step.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

ISO/IEC 42001 requirements for managing third-party and customer relationshipsAdvancedDomain: Artificial intelligence management system (AIMS)

Question 12. A company builds a customer service assistant on top of a large language model provided by an external supplier through an API. How should its AIMS treat this supplier relationship?

  1. Option A: The assistant is out of scope, because the company does not own the model
  2. Option B: The company may only use suppliers that are certified to ISO/IEC 42001
  3. Option C: Signing a contract transfers all accountability for the assistant's outcomes to the supplier
  4. Option D: Allocate responsibilities and manage the supplier in line with its own AI requirements
Show answer and explanation

Correct answer: D. Allocate responsibilities and manage the supplier in line with its own AI requirements

AI systems often depend on models, data and services from others. ISO/IEC 42001 expects the organization to allocate responsibilities clearly and to manage suppliers so that their contributions fit its AI policy and risk treatment.

  • A (incorrect)AI systems that rely on third-party components remain in scope when the organization provides or uses them.
  • B (incorrect)The standard does not require suppliers to hold ISO/IEC 42001 certification.
  • C (incorrect)Contracts can allocate tasks and responsibilities, but the organization remains accountable for its own AIMS and AI use.
  • D (correct)Correct. Third-party and customer relationships are part of the AIMS: responsibilities are allocated and suppliers are managed in line with the organization's requirements.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

First-, second- and third-party AIMS audits, their characteristics and internal audit reporting requirementsFoundationDomain: Artificial intelligence management system (AIMS)

Question 13. A hospital audits the AIMS of the vendor that supplies its AI triage software, using criteria from their contract. What type of audit is this?

  1. Option A: A second-party audit
  2. Option B: A first-party audit
  3. Option C: A third-party audit
  4. Option D: A management review
Show answer and explanation

Correct answer: A. A second-party audit

First party means internal, second party means a customer or other interested party, and third party means an independent organization such as a certification body. ISO/IEC 42001 itself requires internal (first-party) audits at planned intervals.

  • A (correct)Correct. A second-party audit is performed by a party with an interest in the organization, typically a customer auditing its supplier.
  • B (incorrect)First-party audits are internal audits, carried out by or on behalf of the organization on itself.
  • C (incorrect)Third-party audits are performed by independent bodies, such as a certification body, not by a customer.
  • D (incorrect)Management review is top management's periodic review of the AIMS, not an audit by an external customer.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

Corrective action processes and mechanisms for reporting and addressing AI-related concernsIntermediateDomain: Artificial intelligence management system (AIMS)

Question 14. An internal audit finds that a model was deployed without the required impact assessment. The team completes the assessment the same week. What else does ISO/IEC 42001 expect as part of corrective action?

  1. Option A: Nothing, because the immediate fix closes the nonconformity
  2. Option B: A separate preventive action report under a dedicated preventive action clause
  3. Option C: Find the cause, act to prevent recurrence, and check the action worked
  4. Option D: Wait for the next certification audit to decide whether action is needed
Show answer and explanation

Correct answer: C. Find the cause, act to prevent recurrence, and check the action worked

Correction fixes the symptom, corrective action removes the cause. Here, the root cause might be a release process that does not check for a completed impact assessment, so the process needs to change.

  • A (incorrect)Fixing the instance is a correction; corrective action also deals with why it happened.
  • B (incorrect)The harmonized management system structure has no separate preventive action clause; prevention is handled through risk-based planning.
  • C (correct)Correct. Corrective action means evaluating the cause, taking action proportionate to the effects, and reviewing whether it worked, with documented information as evidence.
  • D (incorrect)The organization must react to nonconformities itself; it does not defer the decision to the certification body.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

AI security and operational monitoring, including event logging, performance monitoring, anomaly detection and security event monitoringIntermediateDomain: Artificial intelligence management system (AIMS)

Question 15. Months after deployment, a fraud detection model starts missing more fraud cases because customer behaviour has changed. Which practice in the AIMS is best placed to catch this early?

  1. Option A: Testing the model thoroughly once before deployment and relying on those results
  2. Option B: Event logging and performance monitoring of the deployed system
  3. Option C: Retraining the model on a fixed schedule without measuring performance
  4. Option D: Reviewing the AI policy and AI objectives every three years during the management review
Show answer and explanation

Correct answer: B. Event logging and performance monitoring of the deployed system

The AI system life cycle does not end at deployment. Operation and monitoring, including event logs and performance measures, are what reveal drift, anomalies and security events, and they inform repair, update or decommissioning decisions.

  • A (incorrect)Pre-deployment testing is necessary but cannot detect changes that occur in operation.
  • B (correct)Correct. Monitoring in operation, with logs and performance indicators, detects degradation and anomalies so the organization can act.
  • C (incorrect)Scheduled retraining without measurement may miss or even worsen problems and gives no evidence of performance.
  • D (incorrect)Policy reviews matter but are far too infrequent and high level to detect model degradation.

Reference: PECB: ISO/IEC 42001 Foundation Candidate Handbook (PDF) (opens in a new tab)

Want more practice?

15 questions are a sample. The exam is 40 items.

The full CertRamp ISO/IEC 42001 Foundation practice exams have 240 questions in 6 tests that grow with you, from a diagnostic start to an exam-level Test 4 and a challenge set beyond the exam, with an explanation for every option.

Who writes the exams

Written by a practitioner, not a question mill

Joshua Ravnjak, founder of CertRamp Learning

Joshua Ravnjak

Senior Consultant, digital transformation and AI

Founder, CertRamp Learning

921 learners and 61 reviews on Udemy (as of November 3, 2026)

I have more than 12 years of professional experience and work as a senior consultant in an IT management consultancy, focusing on digital transformation, AI and requirements engineering in the insurance industry. I have earned many certifications myself, so I know what exam preparation feels like. That is why every CertRamp practice exam follows the same progression: an easy start to see where you stand, then real exam level, then beyond it. You walk into the exam with a buffer and a good feeling.

Selected certifications

  • IREB® CPRE Foundation Level
  • ITIL® 4 Foundation
  • Professional Scrum Master I (PSM I)
  • Professional Scrum Product Owner I (PSPO I)
  • Certified SAFe® 6 Scrum Master
  • Certified AI-Empowered SAFe® 6 Product Owner/Product Manager

Plus further certifications from Microsoft and Anthropic (Claude).

ITIL® is a registered trademark of PeopleCert. IREB® is a registered trademark of IREB e.V. SAFe® is a registered trademark of Scaled Agile, Inc. PSM and PSPO are trademarks of Scrum.org.

Also in print

Cover: CPRE Foundation Level Prüfungsvorbereitung 2026 by Joshua Ravnjak

CPRE Foundation Level Prüfungsvorbereitung 2026

6 Übungsprüfungen mit 270 Fragen und ausführlichen Erklärungen

Paperback, German, October 2026

My CPRE Foundation Level exam-prep book (in German) follows the same progressive approach.

View on Amazon (opens in a new tab)