I85PREP
⌘ K
ITIL 4 CAI study materials
ITIL 4 CAI study resources

ITIL 4 CAI practice questions

One question at a time. Build your ITIL 4 CAI understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1AI in service management (scope to be confirmed against official syllabus)

In ITIL 4's service value chain, which activity is most directly supported when an AI chatbot resolves a user's password reset without human involvement?

  1. Plan
  2. Deliver and support Correct answer.
  3. Engage
  4. Improve

Explanation

Deliver and support covers the day-to-day delivery of services to users, including handling incidents and service requests. An automated password reset is a service request fulfilled in that value stream stage, so the chatbot supports deliver and support rather than planning or continual improvement.

Question 2AI in service management (scope to be confirmed against official syllabus)

What does ITIL describe as the primary purpose of the 'incident management' practice?

  1. To define and manage the service catalogue and its offerings
  2. To ensure that changes to services are assessed and authorized
  3. To identify the root cause of all incidents before any restoration work begins
  4. To minimize the negative impact of incidents by restoring normal service operation as quickly as possible Correct answer.

Explanation

Incident management focuses on restoring normal service quickly and limiting negative impact, with workarounds often acceptable. Root cause analysis belongs to problem management, change authorization to change enablement, and catalogue management to service catalogue management.

Question 3AI in service management (scope to be confirmed against official syllabus)

An AI-driven monitoring tool detects a degradation pattern and raises an alert before users report symptoms. Which ITIL 4 concept does this most directly improve?

  1. The 'supplier management' practice
  2. The 'plan' value chain activity
  3. The 'detect' element of the incident management workflow Correct answer.
  4. The 'service financial management' practice

Explanation

Early detection by monitoring shifts incidents from reactive to proactive handling, which is the detect stage of the incident workflow. It does not change planning, supplier relationships, or financial management directly.

Question 4AI in service management (scope to be confirmed against official syllabus)

Which statement best describes 'problem management' in ITIL 4?

  1. It restores service to users as fast as possible through workarounds
  2. It authorizes and schedules changes to production services
  3. It reduces the likelihood and impact of incidents by managing problems and known errors Correct answer.
  4. It records and manages the lifecycle of all service requests

Explanation

Problem management works on the causes of incidents, managing problems and known errors to reduce likelihood and impact. Restoration speed is incident management, authorization is change enablement, and request handling is service request management.

Question 5AI in service management (scope to be confirmed against official syllabus)

A service desk uses an AI assistant that suggests the next best action to an agent during a live chat. Which type of AI capability is being applied?

  1. Sentiment-only analytics
  2. Full automation
  3. Predictive capacity planning
  4. Agent assist Correct answer.

Explanation

Agent assist keeps the human in control while AI surfaces recommendations, knowledge, or next actions. Full automation would remove the human, capacity planning forecasts resources, and sentiment-only analytics is narrower than next-best-action guidance.

Question 6AI in service management (scope to be confirmed against official syllabus)

In ITIL 4, what is a 'known error'?

  1. An incident that has been closed without resolution
  2. A service request that has exceeded its SLA target
  3. A change that has been rejected by the change authority
  4. A problem that has a documented root cause and a workaround Correct answer.

Explanation

A known error is a problem with a documented root cause and, usually, a workaround. It is not a closed incident, a rejected change, or a breached request.

Question 7AI in service management (scope to be confirmed against official syllabus)

Which ITIL 4 guiding principle is most directly reflected when an organization starts AI adoption with a small pilot on one service desk queue before scaling?

  1. Start where you are
  2. Progress iteratively with feedback Correct answer.
  3. Keep it simple and practical
  4. Collaborate and promote visibility

Explanation

Progress iteratively with feedback means breaking work into small, manageable iterations and using feedback to steer the next step. A single-queue pilot that scales based on results fits that principle better than the others.

Question 8AI in service management (scope to be confirmed against official syllabus)

What is the main purpose of the 'service request management' practice?

  1. To investigate the underlying causes of recurring incidents
  2. To manage the lifecycle of all changes to configuration items
  3. To ensure that services are delivered within agreed financial budgets
  4. To support the agreed quality of a service by handling pre-agreed, user-initiated requests Correct answer.

Explanation

Service request management handles user-initiated, pre-agreed requests such as access, information, or standard changes, supporting agreed service quality. Cause investigation, change lifecycle, and budgeting belong to other practices.

Question 9AI in service management (scope to be confirmed against official syllabus)

An organization uses an AI model to classify incoming tickets by category before routing them. Which ITIL 4 dimension does the model itself primarily belong to?

  1. Organizations and people
  2. Value streams and processes
  3. Partners and suppliers
  4. Information and technology Correct answer.

Explanation

The four dimensions are organizations and people, information and technology, partners and suppliers, and value streams and processes. An AI model is a technology and information asset, so it sits in information and technology.

Question 10AI in service management (scope to be confirmed against official syllabus)

A team wants to use AI to draft knowledge articles from resolved tickets. Which risk most directly threatens the quality of the resulting knowledge base?

  1. The model may reduce the number of tickets the service desk receives
  2. The model may shorten the average handling time of agents
  3. The model may consume more storage than the knowledge base allows
  4. The model may reproduce inaccurate or outdated resolutions as authoritative guidance Correct answer.

Explanation

AI-generated content inherits the quality of its source data; if resolutions were wrong or superseded, drafting will propagate errors into the knowledge base. Storage, ticket volume, and handling time are not the primary quality threat described.

Question 11AI in service management (scope to be confirmed against official syllabus)

Which ITIL 4 practice is most directly responsible for defining the utility and warranty of a new AI-enabled service offering?

  1. Relationship management
  2. Service level management
  3. Service catalogue management Correct answer.
  4. Release management

Explanation

Service catalogue management defines and maintains service offerings, including their utility and warranty, so customers know what is available and fit for purpose. Service level management sets targets, relationship management handles stakeholder needs, and release management makes services available.

Question 12AI in service management (scope to be confirmed against official syllabus)

An AI tool predicts which changes are likely to fail based on historical change records. Which ITIL 4 practice does this most directly support?

  1. Incident management
  2. Problem management
  3. Change enablement Correct answer.
  4. Continual improvement

Explanation

Change enablement assesses and authorizes changes, and predictive risk scoring supports that assessment. Incident and problem management handle live issues and causes, while continual improvement addresses broader improvement of value streams.

Question 13AI in service management (scope to be confirmed against official syllabus)

What is the primary role of the 'service desk' practice in ITIL 4?

  1. To manage supplier contracts and performance
  2. To define the technical architecture of new AI services
  3. To capture demand for incident resolution and service requests as the single point of contact between the service provider and users Correct answer.
  4. To authorize changes to production environments

Explanation

The service desk is the entry point and single point of contact for users, capturing demand for incidents and requests. Architecture, supplier contracts, and change authorization belong to other practices.

Question 14AI in service management (scope to be confirmed against official syllabus)

A vendor offers an AIOps platform that ingests telemetry from multiple tools and correlates events. Which ITIL 4 practice does this capability most directly enhance?

  1. Service desk
  2. Monitoring and event management Correct answer.
  3. Capacity and performance management
  4. Problem management

Explanation

Monitoring and event management collects and correlates events from services and components to identify state changes. While correlation can feed problem and capacity work, the described capability is the monitoring and event function itself.

Question 15AI in service management (scope to be confirmed against official syllabus)

Which consideration is most important when deciding whether an AI model's recommendation should be fully automated or require human approval in a service workflow?

  1. The vendor's marketing position on automation
  2. The impact and reversibility of the action the recommendation would trigger Correct answer.
  3. The number of parameters in the underlying model
  4. The programming language used to build the model

Explanation

Automation decisions hinge on the consequence of an error and whether the action can be reversed, which determines the appropriate level of human oversight. Model size, programming language, and marketing claims do not determine risk-appropriate control.

Question 16AI in service management (scope to be confirmed against official syllabus)

What does the ITIL 4 'continual improvement' practice primarily provide?

  1. A register of all configuration items and their relationships
  2. A process for negotiating supplier agreements
  3. A structured approach to identifying and implementing improvements across the organization Correct answer.
  4. A method for restoring service after major incidents

Explanation

Continual improvement provides a structured, ongoing approach to identifying, prioritizing, and implementing improvements. Restoration, configuration records, and supplier negotiation belong to other practices.

Question 17AI in service management (scope to be confirmed against official syllabus)

An AI system used in a regulated service must explain how it reached a decision. Which characteristic of the AI output is being demanded?

  1. Explainability Correct answer.
  2. Scalability
  3. Portability
  4. Latency

Explanation

Explainability is the ability to describe how a model produced an output in terms stakeholders can understand, which regulators often require. Scalability, latency, and portability describe performance and deployment traits rather than interpretability.

Question 18AI in service management (scope to be confirmed against official syllabus)

Which ITIL 4 guiding principle best supports reusing an existing incident classification taxonomy when introducing an AI triage model?

  1. Focus on value
  2. Start where you are Correct answer.
  3. Collaborate and promote visibility
  4. Optimize and automate

Explanation

Start where you are encourages leveraging existing assets and data rather than building anew. Reusing the established taxonomy before automating aligns with that principle and reduces disruption.

Question 19AI in service management (scope to be confirmed against official syllabus)

A service provider wants AI to help forecast demand for a self-service portal so that infrastructure can be scaled ahead of peak periods. Which practice is most directly involved?

  1. Relationship management
  2. Service request management
  3. Incident management
  4. Capacity and performance management Correct answer.

Explanation

Capacity and performance management ensures services meet performance and capacity needs in a cost-effective way, including demand forecasting. Incident and request management handle live demand, and relationship management concerns stakeholder engagement.

Question 20AI in service management (scope to be confirmed against official syllabus)

Which statement best captures the relationship between 'utility' and 'warranty' for an AI-powered service?

  1. Utility and warranty are synonyms describing service availability
  2. Utility is fitness for purpose, while warranty is fitness for use under agreed conditions Correct answer.
  3. Utility describes cost, while warranty describes speed of delivery
  4. Utility is fitness for use, while warranty is fitness for purpose

Explanation

Utility is what the service does, its fitness for purpose, while warranty is how well it performs under agreed conditions, its fitness for use. The two are complementary and both are needed for value.

Question 21AI in service management (scope to be confirmed against official syllabus)

An AI model trained on historical ticket data begins to under-serve a newly onboarded customer segment whose tickets look different from the training data. Which phenomenon is this?

  1. Data drift affecting model relevance Correct answer.
  2. A service request backlog
  3. An incident caused by a failed change
  4. A hardware capacity shortfall

Explanation

When live data diverges from training data, model performance degrades, a form of data drift. The symptom is a relevance and accuracy issue rather than a capacity, change, or backlog problem.

Question 22AI in service management (scope to be confirmed against official syllabus)

In ITIL 4, which practice ensures that accurate and reliable information about the configuration of services and the configuration items that support them is available when and where it is needed?

  1. Service configuration management Correct answer.
  2. Deployment management
  3. Change enablement
  4. Service desk

Explanation

Service configuration management maintains accurate configuration information and relationships between configuration items. The service desk handles contact, change enablement authorizes changes, and deployment management moves components into live environments.

Question 23AI in service management (scope to be confirmed against official syllabus)

A problem manager uses AI to cluster thousands of incidents and identify a latent common cause. Which outcome is the most appropriate first step before raising a known error record?

  1. Immediately publish the cluster as a known error to all users
  2. Validate the suspected common cause with technical investigation and evidence Correct answer.
  3. Escalate the cluster to the change advisory board for immediate approval
  4. Close all clustered incidents as duplicates

Explanation

AI clustering generates hypotheses, not confirmed causes; the problem must be validated before a known error is declared. Publishing unvalidated clusters, mass-closing incidents, or forcing changes would propagate unverified conclusions.

Question 24AI in service management (scope to be confirmed against official syllabus)

Which combination best reflects a responsible approach to using generative AI to draft customer-facing service communications?

  1. Automatic publication to reduce handling time, with corrections handled reactively
  2. Human review of outputs, provenance checks on source content, and a clear escalation path for errors Correct answer.
  3. Restricting the model to internal staff while publishing all outputs unchanged
  4. Disabling logging to protect privacy while allowing unrestricted publication

Explanation

Responsible use pairs automation with human oversight, source verification, and a feedback route to correct errors. Automatic publication without review, unchanged publication, and disabling logging all weaken accountability and quality control.

Question 25AI in service management (scope to be confirmed against official syllabus)

An organization measures AI triage success only by the percentage of tickets routed without human touch. Why is this measure insufficient on its own?

  1. It ignores routing accuracy, downstream rework, and customer outcomes Correct answer.
  2. It conflicts with the definition of a service request
  3. It cannot be collected automatically from ticket systems
  4. It is only valid for incident tickets

Explanation

A touchless rate measures automation coverage, not quality; misrouted tickets create rework, delay, and dissatisfaction. Effective evaluation combines accuracy, rework, resolution time, and customer outcome measures.

Question 26AI in service management (scope to be confirmed against official syllabus)

In ITIL 4, which concept describes the functionality offered by an AI-enabled service and how well it fits the consumer's needs?

  1. Value stream
  2. Service offering
  3. Warranty
  4. Utility Correct answer.

Explanation

Utility is the fitness for purpose of a service: it describes what the service does and whether that functionality supports the consumer's outcomes. Warranty covers fitness for use, such as availability and capacity, so it does not describe functional fit.

Question 27AI in service management (scope to be confirmed against official syllabus)

An AI assistant answers employee HR questions from a curated knowledge base. Which ITIL 4 practice most directly governs the accuracy and lifecycle of that underlying content?

  1. Service desk
  2. Knowledge management Correct answer.
  3. Relationship management
  4. Supplier management

Explanation

Knowledge management is the practice responsible for maintaining and improving the usefulness of information and knowledge across the service value chain, which includes keeping the answers an AI assistant draws on accurate and current. The other practices deal with interaction handling, stakeholder relationships, or suppliers.

Question 28AI in service management (scope to be confirmed against official syllabus)

What is the primary purpose of the 'incident management' practice in ITIL 4?

  1. To identify the root cause of recurring failures
  2. To minimize the negative impact of incidents by restoring normal service operation as quickly as possible Correct answer.
  3. To forecast future demand for services
  4. To define the service catalogue and its dependencies

Explanation

Incident management focuses on restoring normal service operation quickly to limit negative impact. Root cause analysis belongs to problem management, while catalogue definition and demand forecasting sit with other practices.

Question 29AI in service management (scope to be confirmed against official syllabus)

An organization deploys a chatbot that resolves common requests and routes the rest to human agents. Which ITIL 4 dimension does the chatbot's interaction workflow most directly belong to?

  1. Organizations and people
  2. Partners and suppliers
  3. Value streams and processes Correct answer.
  4. Information and technology

Explanation

The workflow of how a chatbot resolves or escalates requests describes how activities are organized and sequenced to create value, which sits in the value streams and processes dimension. Technology is only one component within the information and technology dimension.

Question 30AI in service management (scope to be confirmed against official syllabus)

Which ITIL 4 guiding principle is most directly served by automating a repetitive, well-understood classification task with AI?

  1. Collaborate and promote visibility
  2. Think and work holistically
  3. Focus on value
  4. Optimize and automate Correct answer.

Explanation

Optimize and automate encourages removing manual effort from tasks that are well understood and repeatable, which is exactly what automating a stable classification task achieves. The other principles address value orientation, collaboration, and systemic thinking rather than automation of routine work.

Question 31AI in service management (scope to be confirmed against official syllabus)

A service desk wants AI to suggest the next best action to agents during live chats. Which ITIL 4 practice is most directly enhanced by that capability?

  1. Problem management
  2. Portfolio management
  3. Service desk Correct answer.
  4. Deployment management

Explanation

The service desk practice is the point of contact for users and captures demand, so real-time guidance to agents directly enhances it. Problem management, portfolio management, and deployment management operate on different concerns.

Question 32AI in service management (scope to be confirmed against official syllabus)

What does the ITIL 4 'change enablement' practice ensure when an AI model is used to assess the risk of proposed changes?

  1. That change records are archived permanently
  2. That every change is approved by a senior manager
  3. That changes are deployed without any assessment
  4. That changes are assessed, authorized, and scheduled in a way that maximizes successful service changes Correct answer.

Explanation

Change enablement ensures changes are properly assessed, authorized, and scheduled so that the likelihood of successful change is maximized. AI risk scoring supports this assessment but does not replace authorization or mandate senior approval for every change.

Question 33AI in service management (scope to be confirmed against official syllabus)

Which statement best describes how the 'service value chain' relates to AI initiatives in ITIL 4?

  1. The service value chain applies only to physical infrastructure services
  2. AI initiatives replace the service value chain with a single automated activity
  3. AI can be used only in the deliver and support activity
  4. AI capabilities can support activities such as plan, improve, engage, design and transition, obtain or build, and deliver and support Correct answer.

Explanation

The service value chain is a flexible model of six activities, and AI capabilities can be applied across any of them to help create value. It is not restricted to a single activity or to physical services, nor does it replace the model.

Question 34AI in service management (scope to be confirmed against official syllabus)

An AI model that recommends incident resolutions is trained on two years of historical tickets. Which data-related risk most directly affects the recommendations given to agents?

  1. The model guarantees a lower mean time to restore
  2. The model will always produce the same recommendation regardless of input
  3. The model may reflect biases and gaps present in the historical ticket data Correct answer.
  4. The model eliminates the need for any incident records

Explanation

Recommendations inherit the patterns, biases, and blind spots of the data used to train the model, so historical gaps propagate into its output. Models do not guarantee outcomes, and they still depend on records rather than removing the need for them.

Question 35AI in service management (scope to be confirmed against official syllabus)

A problem manager wants to use AI to correlate monitoring alerts and incident records to surface a shared underlying cause. Which ITIL 4 practice is most directly supported?

  1. Service level management
  2. Service request management
  3. Capacity and performance management
  4. Problem management Correct answer.

Explanation

Problem management identifies and manages the root causes of incidents, and correlating alerts and records to find a shared cause is a core problem-management activity. The other practices address requests, capacity, and service levels rather than root cause identification.

Question 36AI in service management (scope to be confirmed against official syllabus)

A provider plans to let an AI model automatically close incidents that it judges resolved. Which consideration should most directly determine whether full automation is acceptable?

  1. The size of the IT team's office space
  2. The brand color of the self-service portal
  3. The potential impact of an incorrect closure on users and the reversibility of the action Correct answer.
  4. The number of features on the AI vendor's roadmap

Explanation

Whether an action can be safely automated depends on the impact of an error and whether the action can be reversed, which together define the appropriate level of human oversight. Cosmetic, vendor-roadmap, and facilities factors do not bear on the risk of automated decisions.

Question 37AI in service management (scope to be confirmed against official syllabus)

Which ITIL 4 practice is most directly concerned with forecasting the volume of AI-assisted self-service sessions so that compute capacity can be provisioned in advance?

  1. Continual improvement
  2. Service configuration management
  3. Release management
  4. Demand management Correct answer.

Explanation

Demand management understands and influences customer demand for services and helps the provider anticipate the capacity needed to meet it. Release management, configuration management, and continual improvement address different concerns.

Question 38AI in service management (scope to be confirmed against official syllabus)

An AI-enabled service must meet agreed availability targets even though the underlying model is periodically retrained. Which practice most directly ensures those targets are negotiated and monitored?

  1. Incident management
  2. Supplier management
  3. Knowledge management
  4. Service level management Correct answer.

Explanation

Service level management negotiates, agrees, and monitors service levels with customers, which includes availability targets for a service whose model changes over time. The other practices handle incidents, suppliers, and knowledge rather than service-level agreements.

Question 39AI in service management (scope to be confirmed against official syllabus)

A team builds an AI feature to summarize long incident threads for handovers. Which ITIL 4 dimension is most directly engaged by the model, its training data, and the integration with the ticketing tool?

  1. Partners and suppliers
  2. Organizations and people
  3. Information and technology Correct answer.
  4. Value streams and processes

Explanation

The model, the data it uses, and the tool integration are information and technology components. While people and processes are also involved, the question asks specifically about the model, data, and integration.

Question 40AI in service management (scope to be confirmed against official syllabus)

An organization wants AI to help prioritize improvement ideas by estimating their expected value. Which ITIL 4 practice is most directly supported?

  1. Continual improvement Correct answer.
  2. Deployment management
  3. Relationship management
  4. Service desk

Explanation

Continual improvement aligns the organization's practices and services with changing needs through ongoing improvement of products and services, so value-based prioritization of improvement ideas directly supports it. The other practices address interactions, deployment, and stakeholder relationships.

Question 41AI in service management (scope to be confirmed against official syllabus)

Which ITIL 4 guiding principle most directly supports starting AI adoption with a small, reversible pilot before scaling across the organization?

  1. Start where you are
  2. Collaborate and promote visibility
  3. Progress iteratively with feedback Correct answer.
  4. Keep it simple and practical

Explanation

Progress iteratively with feedback encourages organizing work into small, manageable iterations and using feedback to guide the next steps, which describes a pilot-then-scale approach. The other principles address the starting point, simplicity, and collaboration.

Question 42AI in service management (scope to be confirmed against official syllabus)

An AI tool flags configuration items whose recorded attributes appear inconsistent with observed telemetry. Which ITIL 4 practice is most directly supported?

  1. Service request management
  2. Financial management
  3. Workforce and talent management
  4. Service configuration management Correct answer.

Explanation

Service configuration management ensures that accurate and reliable information about configuration items and their relationships is available, so detecting inconsistencies between records and telemetry directly supports it. The other practices address finances, staffing, and requests.

Question 43AI in service management (scope to be confirmed against official syllabus)

Which combination best reflects a responsible approach to using generative AI to draft customer-facing service updates?

  1. Rely on the model's confidence score alone to decide whether to publish
  2. Publish drafts automatically and correct errors only after customers complain
  3. Have a qualified person review and approve drafts before release, with clear ownership of the published content Correct answer.
  4. Disable logging so that drafts cannot be audited

Explanation

Human review and clear ownership keep accountability with the organization while generative AI assists drafting, and it preserves the quality of customer-facing communication. Automatic publishing, confidence-only gating, and disabled logging all weaken control and traceability.

Question 44AI in service management (scope to be confirmed against official syllabus)

A team measures AI triage success only by the share of tickets routed without human touch. Which weakness most directly undermines this metric?

  1. It cannot be reported in a dashboard
  2. It ignores whether routed tickets were handled correctly and resolved for the user Correct answer.
  3. It requires no data collection
  4. It is too expensive to calculate

Explanation

Automation rate measures activity, not outcome; without checking routing accuracy and resolution quality, the metric can reward speed at the expense of correctness. Cost, reporting feasibility, and data collection are not the core weaknesses of the measure.

Question 45AI in service management (scope to be confirmed against official syllabus)

An AI model trained on historical ticket data performs poorly for a newly onboarded customer segment whose tickets look different from the training set. Which action best addresses this under the ITIL 4 principle 'collaborate and promote visibility'?

  1. Work with the new segment to understand their needs and share performance data transparently across teams Correct answer.
  2. Hide the performance gap from the customer to avoid concern
  3. Retrain the model only on the original data to preserve consistency
  4. Remove the segment from the service catalogue

Explanation

Collaboration and visibility means involving stakeholders and sharing relevant information openly, which lets teams understand the new segment's needs and correct the model's blind spot. Concealment, retraining on the same biased data, and removing the segment all avoid the underlying problem.

Question 46AI in service management (scope to be confirmed against official syllabus)

An AIOps platform correlates events from multiple monitoring tools and proposes probable causes. Which ITIL 4 practice benefits most directly from this correlation capability?

  1. Service request management
  2. Monitoring and event management Correct answer.
  3. Portfolio management
  4. Workforce and talent management

Explanation

Monitoring and event management observes services and components and records and reports selected changes of state, so correlating events across tools directly strengthens it. The other practices address requests, investment portfolios, and staffing.

Question 47AI in service management (scope to be confirmed against official syllabus)

A regulated service requires that each AI-assisted decision be explainable and traceable to evidence. Which practice most directly establishes the controls that make this possible?

  1. Capacity and performance management
  2. Information security management Correct answer.
  3. Service desk
  4. Change enablement

Explanation

Information security management protects information and includes controls for confidentiality, integrity, and availability, which encompass traceability and evidence retention for decisions. The other practices address interactions, change authorization, and capacity.

Question 48AI in service management (scope to be confirmed against official syllabus)

A service provider wants AI to decide which incidents should become problems. Which factor most directly determines whether this decision can be delegated to the model?

  1. The font used in the problem register
  2. The model's ability to explain its reasoning and the cost of a missed problem versus a false escalation Correct answer.
  3. The number of meetings held each week
  4. The physical location of the data center

Explanation

Delegating a decision to a model depends on how well its reasoning can be inspected and on the relative cost of the two error types, which determines the acceptable risk. Presentation, meeting frequency, and facility location do not bear on decision delegation.

Question 49AI in service management (scope to be confirmed against official syllabus)

An organization introduces an AI assistant for service desk agents. Which action best reflects the ITIL 4 principle 'think and work holistically'?

  1. Deploy the assistant only to the night shift
  2. Consider how the assistant affects related practices, data flows, and user outcomes across the value chain Correct answer.
  3. Optimize the assistant in isolation without considering its effect on incident, knowledge, and change practices
  4. Measure only the assistant's uptime

Explanation

Thinking and working holistically means understanding how the parts of a system interrelate, so the assistant should be assessed for its effects on connected practices, data, and outcomes. Isolated optimization, partial deployment, and uptime-only measurement all ignore those relationships.

Question 50AI in service management (scope to be confirmed against official syllabus)

A provider must decide between buying a vendor AI service and building an in-house model for ticket classification. Which ITIL 4 practice most directly governs the decision to use an external supplier?

  1. Deployment management
  2. Service desk
  3. Supplier management Correct answer.
  4. Problem management

Explanation

Supplier management ensures that the organization's suppliers and their performance are managed appropriately to support the seamless provision of quality products and services. The service desk, problem management, and deployment management do not govern sourcing decisions.