Governance, Ethics, and Responsible AI in Adoption CAIPM Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A logistics company is deploying an AI system to optimize delivery routes. The system will also predict delivery times for customers. The program manager wants to embed governance across the lifecycle. The team is about to start model training. Which governance activity is most important at this stage?

   • AWait until after deployment to document the model to avoid slowing down the training process.
   • BAsk the legal team to review the model's predictions before training is complete.
   • CFocus only on model accuracy and ignore documentation because the system is low-risk.
   • DDocument the model training process, including hyperparameters and data splits, for reproducibility.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A financial services firm is deploying an AI system for fraud detection. The system will process transactions in real time. The firm must comply with anti-money laundering (AML) regulations. The program manager is conducting a risk assessment. Which risk is most directly related to AML compliance?

   • AThe model may use biased features that discriminate against certain customer groups.
   • BThe model may fail to detect suspicious transactions, leading to regulatory penalties.
   • CThe model may produce false positives that inconvenience legitimate customers.
   • DThe model may be too slow to process transactions in real time.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   Which of the following is a major AI regulation that imposes obligations on organizations based on the risk level of AI systems?

   • AHealth Insurance Portability and Accountability Act (HIPAA)
   • BPayment Card Industry Data Security Standard (PCI DSS)
   • CEU AI Act
   • DGeneral Data Protection Regulation (GDPR)
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   What is the first step in an AI risk management process?

   • ARisk mitigation
   • BRisk monitoring
   • CRisk communication
   • DRisk identification
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A financial institution is deploying an AI system for credit scoring. The model uses alternative data, such as social media activity, to assess creditworthiness. Regulators have expressed concern about potential bias and lack of transparency. The institution wants to maintain its competitive advantage. The program manager must implement risk mitigation. What is the most effective approach?

   • AKeep the alternative data but implement rigorous bias testing and provide explainable AI outputs for each decision.
   • BKeep the model as is and argue that alternative data improves financial inclusion.
   • CUse alternative data only for customers with no traditional credit history, to limit exposure.
   • DEliminate all alternative data from the model to reduce regulatory risk.
   Reveal Solution Discussion Select an answer first