AI Use Case Identification and Value Prioritization CAIPM Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   During prioritization, an AI use case is found to potentially produce biased outcomes for a protected group. Which consideration is most directly relevant?

   • AEthical implications
   • BFeasibility
   • CStrategic alignment
   • DROI
   Reveal Solution Discussion Select an answer first
2. 17expert · hard

   A global bank is prioritizing AI use cases. The European branch wants AI for regulatory compliance, the Asian branch wants AI for customer service, and the North American branch wants AI for fraud detection. The bank has a unified data platform but different regulatory requirements across regions. The CEO wants a single global AI program. What should the AI program manager do?

   • ALet each branch pursue its own use case independently
   • BPrioritize the customer service use case, because it has the highest revenue potential
   • CPrioritize the compliance use case first, because it addresses regulatory risk and can be standardized globally
   • DPrioritize the fraud detection use case, because it is a universal need
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   An insurance company wants to implement an AI system to automate claims processing. The company has a small data science team, limited labeled claims data, and legacy IT systems. Which factor should be the primary consideration when assessing the feasibility of this use case?

   • AAvailability of labeled data and data quality
   • BThe brand reputation of the AI vendor
   • CThe cost of cloud computing resources
   • DThe number of stakeholders interested in the project
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   An insurance company is scoring three AI use cases: (1) claims fraud detection, (2) customer churn prediction, and (3) document processing for underwriting. The scoring model uses four criteria: ROI (30%), strategic alignment (25%), feasibility (25%), and risk (20%). Fraud detection scores high on ROI and strategic alignment but low on feasibility due to data silos. Churn prediction scores moderate on all criteria. Document processing scores high on feasibility and low on risk but moderate on ROI. Which use case should be prioritized based on the weighted scoring model?

   • AChurn prediction, because it has the most balanced scores across all criteria.
   • BDocument processing, because its high feasibility and low risk compensate for moderate ROI, yielding the highest weighted score.
   • CFraud detection, because its high ROI and strategic alignment outweigh the feasibility concerns.
   • DFraud detection and document processing together, because a portfolio approach diversifies risk.
   Reveal Solution Discussion Select an answer first
6. 20expert · hard

   A large e-commerce company is identifying AI use cases. The data science team proposes a recommendation engine, the operations team proposes a demand forecasting model, and the marketing team proposes a customer segmentation tool. The company has abundant transaction data but limited compute resources and a small data science team. The recommendation engine requires significant compute for real-time inference, demand forecasting can run batch jobs, and customer segmentation is a one-time analysis. Which use case should be prioritized first?

   • ADemand forecasting, because it is operationally critical
   • BAll three, because they are independent and can be developed in parallel
   • CCustomer segmentation, because it has low compute requirements and can be delivered quickly with existing data
   • DRecommendation engine, because it has the highest potential revenue impact
   Reveal Solution Discussion Select an answer first