ITIL 4 CAI study plan

Define the scope of AI in service management by writing your own one-page map of where AI touches the ITIL 4 service value chain, marking each activity as assisted, automated, or unaffected.

Distinguish predictive, generative, and rules-based AI well enough to classify ten real service desk or operations use cases without looking anything up.

Explain how AI changes the demand, value, and outcome elements of the ITIL 4 service value system in your own words, using one worked example from incident handling.

Identify the roles most exposed to AI augmentation (service desk analyst, problem manager, change approver) and note what each role keeps doing that AI does not.

Trace a single incident from user report to closure and mark every step where AI could contribute, citing the specific practice that owns each step.

Compare how AI affects three practices: incident management, problem management, and service desk, listing the inputs and outputs that change.

Draft candidate metrics for AI-assisted incident resolution (for example deflection rate, mean time to restore, first-contact resolution) and state what each metric can and cannot prove.

Explain why AI use in a practice must still respect that practice's purpose and guiding principles rather than replacing them.

Map AI contributions across the full service value chain, identifying at least two intervention points per activity and the practice that governs each.

Design an AI-assisted change enablement flow that keeps risk assessment and authorization decisions with accountable humans.

Describe how AI can support continual improvement by surfacing patterns in incident, problem, and feedback data, and specify what evidence a human needs before acting.

Identify the workflow data an AI tool would need for your chosen use case and note where that data lives and who owns it.

List the principal AI risks in a service context (hallucination, bias, data leakage, automation bias, model drift, vendor dependency) and match each to a plausible service failure.

Explain how existing ITIL 4 governance and risk management concepts apply to AI systems used in service delivery.

Draft a simple risk register with five AI risks, each with likelihood, impact, owner, and a mitigation that a service organization could actually run.

Distinguish accountability from responsibility when an AI-assisted decision causes harm, and state who answers for it.

Draft a proportionate AI governance policy covering approved use cases, prohibited uses, human oversight, and review cadence.

Define the human-in-the-loop checkpoints for one AI-assisted process, specifying what the human reviews and what they are authorized to override.

Apply relevant regulatory and ethical expectations (transparency, explainability, data protection, fairness) to your chosen use case without overstating legal certainty.

Explain how AI changes the information security and supplier management practices, including third-party model providers.

Identify the ethical tensions in an AI service decision (fairness across user groups, transparency to customers, worker impact) and argue both sides of one.

Design a feedback mechanism that captures user and staff experience with AI-assisted services and routes it into continual improvement.

Explain how automation bias and over-trust can degrade service quality, and propose a countermeasure that fits normal operations.

Assess how AI affects the workforce: which tasks shift, what reskilling is needed, and how to communicate the change honestly.

Integrate your week-by-week artifacts into a single coherent AI adoption proposal for one service organization.

Reconcile conflicts between your governance policy, risk register, and value stream design so the documents agree on owners and controls.

Identify and close your three weakest areas by writing short explanations of each from memory, then checking them against ITIL 4 concepts.

Explain how AI adoption itself should be run as a continual improvement initiative with measurable outcomes.

Answer practice questions across all four domains under timed conditions and record your accuracy per domain.

Rework any domain scoring below your target by rewriting the underlying concepts rather than re-reading notes.

Summarize each domain in a one-page memory sheet built only from your own words and examples.

Verify your exam logistics and confirm which syllabus areas remain unconfirmed so you know where to keep checking.