PECB Certified Artificial Intelligence Professional study plan

Define artificial intelligence, machine learning, deep learning, and generative AI, and distinguish them with one original example each.

Classify the main ML paradigms (supervised, unsupervised, reinforcement, self-supervised) and state one realistic business use case per paradigm.

Explain the standard ML lifecycle from problem framing through data preparation, training, evaluation, deployment, and monitoring.

Identify the roles in an AI project team and describe the decision rights of each.

Compare precision, recall, F1, ROC-AUC, and confusion-matrix interpretation, and select the right metric for three imbalanced-class scenarios.

Diagnose overfitting and underfitting from described training and validation behavior, and propose a concrete remedy for each.

Explain how foundation models, transformers, and prompt-based interaction change the build-versus-buy decision for an organization.

Describe human oversight mechanisms for AI outputs, including human-in-the-loop and human-on-the-loop patterns.

Define AI risk and differentiate it from general IT risk using at least four AI-specific characteristics such as opacity, autonomy, and data dependence.

Build a risk register containing at least ten AI risks mapped to likelihood, impact, and an owner.

Apply a recognized risk treatment logic (avoid, mitigate, transfer, accept) to five of the registered risks.

Explain how risk appetite statements translate into concrete AI project thresholds.

Describe the structure and intent of the leading AI risk management frameworks and explain how they complement rather than replace enterprise risk management.

Map bias, drift, and model degradation to detection methods and monitoring cadence.

Draft acceptance criteria for a third-party AI component covering performance, transparency, and audit rights.

Outline an AI incident response process from detection through containment, notification, and post-incident review.

Distinguish AI governance from AI management and place both within an organization's existing corporate governance structure.

Draft an AI policy outline covering acceptable use, accountability, approval gates, and prohibited uses.

Assign accountability using a responsibility matrix across executive, technical, legal, and business functions.

Explain how AI governance requirements drive or constrain digital transformation initiatives.

Describe how AI capabilities reshape operating models, and identify three transformation use cases with measurable business outcomes.

Explain transparency, explainability, and contestability obligations toward affected stakeholders and how to operationalize them.

Identify the main data protection principles that apply to AI training and inference, including purpose limitation, minimization, and retention.

Compare anonymization, pseudonymization, and synthetic data as safeguards, stating the residual re-identification risk of each.

Map the AI attack surface, covering data poisoning, adversarial input, model extraction, prompt injection, and insecure output handling.

Select controls for each identified threat and state where in the lifecycle the control applies.

Explain secure design expectations for AI systems, including access control, logging, and supply chain assurance for models and datasets.

Describe how to conduct a privacy impact assessment for an AI system and what triggers one.

Rehearse scenario-based questions across all four domains under timed conditions and review every error against the source concept.

Reconcile your week 3 risk register and week 5 policy outline into a single consistent governance narrative.

Explain how security and data protection controls feed back into risk treatment and governance reporting.

Close remaining knowledge gaps by rewriting weak notes into short, exam-ready summaries.