AI Strategy Frameworks CAIPM Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   An energy company applies an AI strategy framework to prioritize use cases. The framework requires each use case to have a clear owner, a defined success metric, and an identified data source. A proposed use case for predictive maintenance on wind turbines has a clear owner and success metric but the data source is a legacy SCADA system that is not yet integrated. What should the program manager do?

   • AReplace the legacy SCADA data with synthetic data to satisfy the framework requirement.
   • BInclude the use case in the portfolio but flag it as 'data-dependent' and create an integration plan.
   • CReject the use case because the data source is not currently integrated.
   • DMark the use case as ready and address data integration during implementation.
   Reveal Solution Discussion Select an answer first
2. 27application · medium

   A regional bank with a mature IT organization and a centralized decision-making culture wants to adopt AI for fraud detection and customer service automation. The program manager must select a framework that provides clear governance, aligns with existing enterprise architecture, and accommodates the bank's risk-averse compliance requirements. Which framework selection is most appropriate?

   • AAdopt a lightweight, bottom-up experimentation framework to encourage grassroots innovation across business units.
   • BChoose a framework designed for startups with minimal process overhead to accelerate time-to-market for AI features.
   • CSelect a framework focused solely on algorithmic fairness metrics, ignoring deployment and infrastructure considerations.
   • DUse a top-down, stage-gated framework that integrates with enterprise architecture and includes compliance checkpoints at each phase.
   Reveal Solution Discussion Select an answer first
4. 28expert · hard

   A hospital network applies an AI strategy framework to its clinical AI portfolio. The framework includes a 'clinical safety' gate that requires evidence of model validation before deployment. A use case for sepsis prediction has strong preliminary results but has not been validated on the hospital's patient population. The clinical team wants to deploy immediately to save lives. What should the program manager do?

   • ABypass the clinical safety gate and document the decision as an exception.
   • BDeploy the model immediately because the potential to save lives outweighs the validation requirement.
   • CApprove a limited, monitored pilot with strict oversight while validation on the hospital's population is completed.
   • DReject the use case because it has not been validated on the hospital's population.
   Reveal Solution Discussion Select an answer first
5. 29foundation · easy

   When selecting an AI strategy framework, which factor should be considered to ensure the framework fits the organization's current state?

   • AThe organization's AI maturity level
   • BThe brand of the organization's cloud provider
   • CThe programming language used for legacy systems
   • DThe number of employees in the IT department
   Reveal Solution Discussion Select an answer first
6. 30application · medium

   A healthcare provider is applying the AI Strategy Framework to identify use cases. The framework includes a 'stakeholder impact assessment' component. Which activity best demonstrates this component?

   • ASurveying clinicians and patients about potential AI tools
   • BReviewing the organization's data storage capacity
   • CCalculating the ROI of each AI use case
   • DComparing AI vendors' pricing models
   Reveal Solution Discussion Select an answer first