Identify the Foundational Concepts of Generative AI AB-731 Practice Questions (Page 5)

Questions 21–25

1. 21expert · hard

   A logistics company wants to use AI to optimize delivery routes. They have historical delivery data. They are considering generative AI. What is the most appropriate solution?

   • AUse a generative AI model to generate new delivery routes based on historical data.
   • BUse a generative AI model to write reports about delivery performance.
   • CUse a generative AI model to replace the need for historical data.
   • DUse a traditional optimization algorithm to find the most efficient routes.
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A financial advisory firm wants to use generative AI to produce investment summaries for clients. They are worried about the AI generating confident but incorrect statements about market trends. What is the most effective way to reduce this risk?

   • AUse a larger model with more parameters to ensure factual accuracy.
   • BLimit the AI to only use data from the last 24 hours.
   • CGround the AI's responses in a curated set of approved financial data and require citations for all claims.
   • DAsk the AI to include a disclaimer that the information may not be accurate.
   Reveal Solution Discussion Select an answer first
4. 23expert · hard

   A software development company wants to use generative AI to automatically generate code documentation from source code. They have a large codebase and need the documentation to be accurate and consistent. They have a limited budget. What is the best approach?

   • AFine-tune a smaller model on a sample of well-documented code to generate documentation for the entire codebase.
   • BUse a pretrained model with zero-shot prompting to generate documentation without any fine-tuning.
   • CUse a large pretrained model with detailed prompts to generate documentation for each function.
   • DUse a traditional documentation generator that extracts comments and signatures.
   Reveal Solution Discussion Select an answer first
5. 24expert · medium

   A government agency wants to use generative AI to draft responses to citizen inquiries. They have a strict requirement that all responses must be factually accurate and free of bias. They also have a limited budget and cannot afford extensive human review for every response. Which approach best meets these conflicting requirements?

   • AUse a pretrained model with RAG that retrieves information from approved government databases, and implement a sampling-based human review process.
   • BUse a fine-tuned model on government documents and rely on the model's training to ensure accuracy.
   • CUse a traditional AI model to classify inquiries and provide template responses.
   • DUse a large pretrained model without any grounding, assuming it is accurate enough.
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   A company wants to automatically generate personalized product descriptions for thousands of items. Which type of AI approach is best suited for this task?

   • ATraditional AI regression model
   • BTraditional AI classification model
   • CTraditional AI clustering model
   • DGenerative AI language model
   Reveal Solution Discussion Select an answer first