Identify the Foundational Concepts of Generative AI AB-731 Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A retail company wants to use generative AI to create personalized marketing emails for customers. They have a large customer database with purchase history. They need the emails to be highly relevant and engaging. What is the best approach?

   • AFine-tune a model on past successful marketing emails and use customer data in prompts.
   • BUse a rule-based system that inserts customer names and product recommendations into templates.
   • CUse a pretrained model with prompts that include customer purchase history.
   • DUse a traditional recommendation engine to suggest products and generate emails manually.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A software company is evaluating whether to use a generative AI API to auto-generate code comments. They estimate that generating comments for their codebase will consume 10 million tokens per month. The API costs $0.02 per 1,000 tokens. The company currently spends $5,000 per month on manual code review. What is the most important consideration for the ROI analysis?

   • AThe token cost is irrelevant because the company can fine-tune a model for free.
   • BThe company should only consider the cost of the API, not the benefits.
   • CThe API cost is too high compared to the manual review cost, so it is not worth it.
   • DCompare the API cost ($200/month) with the time saved by developers, not just the review cost.
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A customer support organization wants to deploy a generative AI assistant to draft responses to common queries. They expect high volume but have a limited budget. They need to minimize cost while maintaining quality. Which strategy should they adopt?

   • ADeploy multiple large models in parallel and select the best response via a voting mechanism.
   • BFine-tune a smaller model on historical support tickets to generate responses, reducing token usage per query.
   • CUse a large pretrained model with extensive prompt engineering to handle all queries with maximum accuracy.
   • DUse a traditional chatbot with predefined responses to avoid generative AI costs entirely.
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   In which scenario can generative AI deliver significant business value through automation?

   • AReplacing all human employees in the organization
   • BGuaranteeing 100% accuracy in all outputs
   • CEliminating the need for data storage
   • DAutomating the creation of routine customer support responses
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A manufacturing company wants to use AI to generate safety inspection reports from sensor data. The reports must follow a strict regulatory format and include specific data points. The company has a large dataset of past reports. Which approach is most suitable?

   • AFine-tune a pretrained generative model on the historical reports to learn the required format and terminology.
   • BUse a traditional AI model to classify sensor data as safe or unsafe and generate a simple text report.
   • CUse a pretrained model with generic prompts and hope it follows the regulatory format.
   • DUse a pretrained model without any customization because it already knows all regulations.
   Reveal Solution Discussion Select an answer first