mediumMultiple Choice
Generative AI Leader Practice Question: Using a large language model for a…
A company is using a large language model for a customer-facing chat application. They notice that the model sometimes generates plausible-sounding but incorrect information. Which strategy is most effective to reduce this issue?
⚠ Common exam trap
Generative AI Leader often tests the misconception that fine-tuning eliminates hallucination — candidates pick fine-tuning when the real fix is grounding via RAG, and they confuse temperature direction (higher = more random).
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
✓
Implement Retrieval-Augmented Generation (RAG) to retrieve relevant documents before generating a response.
Retrieval-Augmented Generation (RAG) grounds the model's responses in retrieved, authoritative documents at inference time, which directly reduces hallucination by supplying factual context the model can cite. It does not require retraining and can be updated as source documents change. This is the most effective and operationally practical mitigation for plausible-sounding but incorrect outputs.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Fine-tune the model on a dataset of corrected conversations.
Why it's wrong here
Fine-tuning on corrected conversations teaches style and format, not factual grounding; the model still hallucinates on unseen queries. Fine-tuning suits adapting tone, domain vocabulary or task structure, whereas retrieval-augmented generation supplies authoritative source content at inference time, which is what this scenario requires.
- ✗
Reduce the context window to limit the information the model considers.
Why it's wrong here
Shrinking the context window removes the grounding material the model needs, so it has less evidence to condition on and hallucinates at least as often. It is tempting because a smaller window cuts cost and latency, but retrieval-augmented generation with cited sources addresses fabrication directly.
- ✓
Implement Retrieval-Augmented Generation (RAG) to retrieve relevant documents before generating a response.
Why this is correct
Retrieval-Augmented Generation grounds responses in retrieved, verifiable documents rather than relying solely on parametric memory, directly reducing hallucinated content. By supplying relevant source passages at inference time, the model cites factual context instead of inventing plausible details, satisfying the stem's requirement to curb fabricated but convincing answers in the customer-facing chat.
- ✗
Increase the temperature parameter to make the model more deterministic.
Why it's wrong here
Raising temperature increases sampling randomness, which worsens hallucination rather than reducing it; determinism comes from lowering temperature toward zero. Temperature tuning is genuinely useful for controlling creativity and diversity in open-ended generation tasks, but it cannot ground outputs in verified facts.
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 and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Google Cloud exam blueprint
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.