Establishing Organizational Alignment for AI CAIPM Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   An e-commerce company has deployed an AI recommendation system. Six months later, the marketing team reports that click-through rates have improved, but the customer service team reports an increase in complaints about irrelevant recommendations. The program manager needs to ensure continuous alignment. What is the most appropriate action?

   • AEstablish a feedback loop that includes both quantitative KPIs and qualitative feedback from customer service to adjust the AI model.
   • BTransfer the complaint data to the marketing team for them to handle.
   • CDiscontinue the AI system because of the complaints.
   • DRely solely on the click-through rate as the success metric since it is the primary KPI.
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A government agency has deployed an AI system for citizen service requests. Six months after launch, usage is lower than expected, and some departments are bypassing the system. The program manager needs to ensure continuous alignment. What is the best action?

   • AConduct a one-time survey to measure user satisfaction and then move on.
   • BMandate that all departments use the system and penalize non-compliance.
   • CRe-launch the same communication campaign that was used at the initial launch.
   • DSet up a quarterly review process that collects feedback from users and department heads, and adjusts the AI system and communication strategies accordingly.
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A government agency has deployed an AI system for fraud detection. The system has been successful in identifying fraudulent claims, but it has also flagged many legitimate claims, causing delays and complaints. The program manager needs to ensure continuous alignment. What is the best action?

   • ADisable the AI system and return to manual fraud detection to eliminate false positives.
   • BKeep the AI system as is, since it is catching fraudulent claims, and focus on improving the appeals process.
   • CIncrease the AI system's sensitivity to catch more fraudulent claims, even if it means more false positives.
   • DEstablish a feedback mechanism that collects data on false positives and claimant complaints, and use that data to adjust the AI model's thresholds and improve the claims review process.
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A hospital is deploying an AI system to predict patient readmission risk. The nursing staff is concerned about extra documentation, while the finance department wants to reduce readmission penalties. The medical board is skeptical about the AI's clinical validity. The program manager needs to build cross-functional support. What is the most effective approach?

   • ACreate a multidisciplinary steering committee with nursing, finance, and medical board representatives to oversee the AI deployment.
   • BFocus only on the finance department's goal of reducing penalties, since it is the primary driver.
   • CHire an external consultant to manage the deployment without involving internal stakeholders.
   • DImplement the AI system first and then address concerns after it proves its value.
   Reveal Solution Discussion Select an answer first
6. 5expert · hard

   A hospital network is implementing an AI diagnostic support system. The radiologists are concerned about liability and want to review all AI suggestions, while the emergency department wants faster turnaround times and is willing to accept AI recommendations without review. The legal department requires clear accountability for diagnostic decisions. The program manager must build cross-functional support. What is the most effective approach?

   • AEstablish a governance committee with representatives from radiology, emergency, and legal to define a clear escalation and accountability framework.
   • BLet each department decide its own policy for AI use independently.
   • CRequire radiologists to review every AI suggestion, even in emergency cases, to ensure safety.
   • DAllow the emergency department to use AI recommendations without review to maximize speed.
   Reveal Solution Discussion Select an answer first