3.1 Describe How to Proactively Overcome Foundation Model Limitations. GENERATIVE-AI-LEADER Practice Questions (Page 5)

Questions 21–22

1. 21application · medium

   A financial services firm is deploying a generative AI assistant to summarize internal policy documents. The compliance team requires that every summary be traceable to a specific source paragraph. The model currently produces fluent summaries but occasionally invents details. Which mitigation should the team implement first?

   • AFine-tune the model on the internal policy documents so it memorizes the content.
   • BUse grounding with the internal document repository and require citations in the prompt.
   • CDeploy the model with a higher temperature setting to increase creativity in summaries.
   • DRestrict the model's input token limit to force it to focus on shorter text.
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A medical research team is using a foundation model to summarize clinical trial reports. The model sometimes omits critical safety information. The team has a strict requirement that no safety data be missed. Which approach best ensures completeness?

   • AFine-tune the model on all clinical trial reports to improve its summarization skills.
   • BUse a human-in-the-loop process where a clinician reviews and verifies every summary.
   • CIncrease the model's max output tokens to force it to include more details.
   • DUse a lower temperature to make the model more focused on the input text.
   Reveal Solution Discussion Select an answer first