GENERATIVE-AI-ENGINEER-ASSOCIATE · Question #44
A Generative AI Engineer has developed an LLM application to answer questions about internal company policies. The Generative AI Engineer must ensure that the application doesn't hallucinate or leak c
Sign in or unlock GENERATIVE-AI-ENGINEER-ASSOCIATE to reveal the answer and full explanation for question #44. The question stem and answer options stay visible for context.
Question
A Generative AI Engineer has developed an LLM application to answer questions about internal company policies. The Generative AI Engineer must ensure that the application doesn’t hallucinate or leak confidential data. Which approach should NOT be used to mitigate hallucination or confidential data leakage?
Options
- AAdd guardrails to filter outputs from the LLM before it is shown to the user
- BFine-tune the model on your data, hoping it will learn what is appropriate and not
- CLimit the data available based on the user's access level
- DUse a strong system prompt to ensure the model aligns with your needs.
Unlock GENERATIVE-AI-ENGINEER-ASSOCIATE to see the answer
You've previewed enough free GENERATIVE-AI-ENGINEER-ASSOCIATE questions. Unlock GENERATIVE-AI-ENGINEER-ASSOCIATE for full answers, explanations, the timed quiz mode, progress tracking, and the master PDF. Question stem and options stay visible so you can still see what's on the exam.