Types of AI - ML, DL, Generative AI, and Agents CAIPM Practice Questions (Page 6)

Questions 26–30

1. 26expert · hard

   A legal firm wants to automate the initial review of contracts to identify clauses that deviate from standard templates. They have a small set of labeled examples (500 contracts) and need the system to explain why a clause is flagged. The firm also wants to generate a summary of each contract for attorneys. Which approach best balances accuracy, explainability, and content generation?

   • AUse a generative AI model alone to both classify and summarize contracts.
   • BUse an AI agent to autonomously negotiate contract terms with the other party.
   • CUse a single deep learning model that performs both classification and summarization.
   • DUse a supervised ML model for clause classification with explainable features, and a separate generative AI model for contract summaries.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   A marketing team wants to automatically produce personalized product descriptions for thousands of items. Which AI capability is most directly applicable?

   • AReinforcement learning to optimize pricing strategies
   • BGenerative AI to write unique descriptions based on product attributes
   • CPredictive analytics to forecast sales for each product
   • DClustering algorithms to group similar products
   Reveal Solution Discussion Select an answer first
4. 28expert · hard

   A large e-commerce company wants to automate customer support. The system must handle common queries, escalate complex issues, and also generate personalized product recommendations. The company has a vast amount of historical chat logs and purchase data. They want to minimize development effort while maintaining high accuracy. What approach should they take?

   • ATrain a single deep learning model from scratch on chat logs and purchase data to handle everything
   • BUse a rule-based system with predefined responses and manual recommendation logic
   • CBuild an AI agent that uses a pre-trained generative model for responses and a recommendation ML model, with escalation rules
   • DDeploy a generative model alone to generate responses and recommendations without an agent
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A manufacturing company wants to predict equipment failure before it happens using sensor data (temperature, vibration, pressure). They have a moderate amount of labeled failure data and need a model that can be updated frequently as new sensor data arrives. Which approach is most suitable?

   • ADeploy a generative AI model to create synthetic sensor data for analysis.
   • BUse an AI agent to autonomously replace equipment based on a schedule.
   • CUse a deep learning model that requires a very large dataset to train effectively.
   • DTrain a supervised machine learning model (e.g., random forest) on the sensor features to classify normal vs. failure conditions.
   Reveal Solution Discussion Select an answer first
6. 30foundation · easy

   Which business application is best suited for deep learning rather than traditional machine learning?

   • ARecognizing objects in real-time video streams from security cameras
   • BPredicting loan default using a small tabular dataset with clear features
   • CSorting customer emails into predefined folders using simple keyword rules
   • DCalculating monthly sales totals from a spreadsheet
   Reveal Solution Discussion Select an answer first