AI-102 Implement generative AI solutions Practice Question
You deploy a GPT-4o model in Azure OpenAI Service for an internal assistant that summarizes long contracts. Legal requires that every generated summary be traceable to the exact source passages and that the assistant never answer from general world knowledge. You need the model to ground each statement in retrieved text and expose the supporting passages to the caller. What should you configure?
⚠ Common exam trap
The trap here is believing that lowering temperature removes hallucination and therefore substitutes for retrieval-based grounding.
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
✓
Use the Azure OpenAI On Your Own Data feature with citations enabled
Grounding the assistant in an indexed corpus and returning citations is what makes each summary statement traceable to a specific contract passage. Azure OpenAI On Your Own Data performs retrieval against Azure AI Search, injects the retrieved chunks, and returns citation metadata with the response. Sampling parameters, output length, and monitoring logs change behavior or observability but never bind statements to sources.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Enable the abuse monitoring logging option on the Azure OpenAI resource
Why it's wrong here
Abuse monitoring stores prompts and completions for review of harmful content, supporting compliance investigations. It does not constrain the model to retrieved passages or generate citations in responses. Logging records what was asked and answered rather than grounding answers in source text, so it does not meet the legal traceability requirement in this scenario.
- ✗
Increase the model's max_tokens value to fit the entire contract
Why it's wrong here
Raising max_tokens allows a longer completion, but it does not change where the model draws information. Even with the full contract in the prompt, the model may paraphrase from pretraining and cannot reliably emit passage-level citations. The parameter governs output length, so it fails to deliver the required traceability and grounding behavior.
- ✗
Set temperature to 0 and top_p to 1 on the chat completions call
Why it's wrong here
Low temperature and top_p make sampling more deterministic, which stabilizes wording, but they do not force the model to cite retrieved passages or restrict it to them. The model can still produce fluent statements drawn from pretraining. Determinism improves reproducibility of a summary, not its traceability, so this setting alone cannot satisfy the legal grounding requirement.
- ✓
Use the Azure OpenAI On Your Own Data feature with citations enabled
Why this is correct
On Your Own Data connects the deployment to an Azure AI Search index and instructs the model to answer only from retrieved chunks, returning citations that map statements to source passages. This directly provides the traceability legal requires and constrains responses to the indexed contracts rather than general knowledge. It is the supported grounding pattern for this scenario.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI-102 question from scratch — 761 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or 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 Microsoft exam blueprint
This AI-102 practice question is part of Courseiva's free Microsoft 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 AI-102 exam.