Core AI Concepts and Business Applications CAIPM Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A marketing team wants to use AI to personalize email campaigns. They have a large dataset of customer purchase history and click behavior. A data scientist suggests using a deep learning model, while a colleague suggests a simpler machine learning model. The team has limited computational resources and needs to deploy the solution quickly. What is the most appropriate consideration when choosing between deep learning and traditional machine learning?

   • ATraditional machine learning may be sufficient and more practical given the resource constraints and deployment timeline.
   • BDeep learning is the only option that can handle large datasets.
   • CDeep learning always outperforms traditional machine learning on tabular data.
   • DDeep learning does not require any data preprocessing.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A financial services firm wants to use AI to approve small business loans. They have historical loan data that includes applicant demographics, financial history, and loan outcomes. The compliance team is concerned about regulatory requirements for fair lending. What is the most important limitation to address?

   • AThe AI model may inadvertently discriminate against protected groups if the training data contains historical biases.
   • BThe AI model will automatically ensure compliance with fair lending laws.
   • CThe AI model will eliminate the need for human loan officers entirely.
   • DThe AI model will be unable to process numerical financial data.
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A regional bank wants to reduce fraudulent credit-card transactions. The fraud team currently uses a set of manually written rules (e.g., flag transactions over $500 from a new device). The bank's data science team has access to millions of labeled historical transactions and wants to deploy a system that learns complex patterns automatically and improves as new fraud cases are confirmed. Which approach best fits this business need?

   • AExpand the manual rule set to cover more device and amount combinations.
   • BUse a simple linear regression model to predict the dollar amount of each transaction.
   • CTrain a deep learning model on historical transaction data and integrate it into the real-time authorization workflow.
   • DDeploy a chatbot to ask customers to confirm suspicious transactions before processing.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A hospital wants to use AI to predict patient readmission risk. The data science team has access to electronic health records (EHRs) from the last 10 years, but the hospital's patient population has changed significantly over that period (e.g., new demographics, new treatments). The model will be used to allocate care-management resources. What is the most important limitation to address before deployment?

   • AThe model may not generalize to the current patient population due to data drift.
   • BThe model will replace the need for human clinical judgment entirely.
   • CThe model will automatically comply with all healthcare regulations without review.
   • DThe model will be unable to process structured EHR data.
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A manufacturing company wants to implement AI to optimize energy consumption in its factories. The company has a limited budget and the AI must integrate with existing legacy control systems. The plant managers are skeptical of AI and require that any recommendation be understandable. What is the best way to demonstrate value and gain buy-in?

   • ADeploy a complex deep learning model across all factories immediately.
   • BPurchase a commercial AI solution and integrate it without testing.
   • CFocus on theoretical benefits and avoid implementation until the budget increases.
   • DRun a pilot project in one factory using a simple, interpretable model and present the energy savings to stakeholders.
   Reveal Solution Discussion Select an answer first