mediumMultiple Choice
Generative AI Leader Practice Question: During a proof-of-concept, a team notices that…
During a proof-of-concept, a team notices that their GenAI-powered meeting summarizer occasionally includes hallucinated details. They want to improve summary accuracy before production. Which action would be most effective?
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Provide the meeting transcript as context in the prompt
Providing the meeting transcript as context grounds the model in factual data, reducing hallucinations. Few-shot prompting may help, but the primary issue is lack of source material. Adjusting temperature is less impactful. Switching to a larger model without context may not help.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Upgrade to a larger model
Why it's wrong here
A larger model may reduce some errors but does not guarantee factual grounding; it can still hallucinate details absent from the transcript. Larger models suit complex reasoning or nuanced generation tasks. The scenario needs retrieval-augmented generation or grounding the summary in the source transcript.
- ✓
Provide the meeting transcript as context in the prompt
Why this is correct
Grounding the model in the actual meeting transcript supplies the factual source material, so summaries are generated from retrieved content rather than parametric memory. This retrieval-augmented approach directly reduces fabricated details, satisfying the accuracy constraint better than prompt wording tweaks.
- ✗
Decrease the temperature to 0.0
Why it's wrong here
Lowering temperature sharpens token sampling but cannot inject facts absent from the context; hallucinated meeting details persist because the model still fabricates unsupported content. Temperature tuning suits creative variation control, not grounding. Retrieval-augmented generation or grounding the prompt in the transcript addresses accuracy directly.
- ✗
Use few-shot prompting with example summaries
Why it's wrong here
Few-shot examples shape output style and format but do not ground the model in the meeting transcript, so fabricated details persist. It is tempting because prompting is quick to change, and it would help when the task is formatting or tone rather than factual accuracy.
Go deeper
Related to this question
About these practice questions
This Generative AI Leader question is part of Courseiva's 1,008-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This Generative AI Leader practice question is part of Courseiva's free Google Cloud certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the Generative AI Leader exam.