AI-900 Practice Question: Describe features of generative AI workloads on Azure
A company wants to build a chatbot that answers customer questions using a large language model. The company has an extensive internal knowledge base with accurate, up-to-date product information. To ensure the chatbot's answers are based on this reliable source rather than the model's internal knowledge, which technique should they use?
⚠ Common exam trap
A common mix-up: candidates confuse fine-tuning (which alters the model's internal knowledge) with retrieval-augmented generation (which keeps the model unchanged and instead supplies external context at query time), leading them to incorrectly select fine-tuning as the method to ensure answers come from a specific knowledge base.
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
✓
Grounding with retrieval-augmented generation
Grounding with retrieval-augmented generation (RAG) retrieves relevant, up-to-date chunks from the internal knowledge base and provides them as context to the large language model (LLM) at inference time. This ensures the chatbot's answers are factually based on the company's reliable source rather than relying on the model's potentially outdated or incorrect parametric memory.
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-tuning the model on the knowledge base
Why it's wrong here
Fine-tuning adjusts the model's weights on curated question-answer pairs or documents from the knowledge base, which can embed some domain knowledge. However, this only bakes in a static snapshot: the model cannot query the live knowledge base, and any update to the documents requires full retraining. It also risks memorized or distorted facts without grounding or source citations, making it unsuitable for a chatbot that must answer accurately from a changing knowledge base.
- ✗
Zero-shot learning
Why it's wrong here
Zero-shot learning means the model is asked to answer with no task-specific examples and no external data access, relying entirely on knowledge captured during pretraining. Because the target knowledge base is neither consulted nor injected into the prompt, the chatbot cannot retrieve current product details, pricing, or policies from it. This approach may work for general questions but fails the requirement that answers be grounded in, and verifiable against, the organization's knowledge base.
- ✓
Grounding with retrieval-augmented generation
Why this is correct
Grounding with retrieval-augmented generation (RAG) solves the problem by vectorizing and indexing the knowledge base, retrieving the most relevant passages for the user's question, and injecting those passages into the model's prompt as context. The model then generates an answer conditioned on that retrieved evidence, which means responses reflect the current KB contents and can be cited back to source documents. Updates to the KB only require re-indexing the changed documents, not retraining the model, and the approach dramatically reduces hallucination compared with using the model's internal knowledge alone.
- ✗
Prompt engineering with few-shot examples
Why it's wrong here
Few-shot prompting supplies the model with a handful of input-output examples to demonstrate the desired response style, tone, or output format. It does not retrieve or insert any content from the knowledge base, so the model must still rely on its own parametric memory for factual details. That makes it a useful technique for shaping the chatbot's behavior, but it cannot ensure the answer matches the latest, authoritative information in the knowledge base.
Go deeper
Related to this question
Learn chapter
Azure AI Language Service
Key term
RAG
Retrieval-Augmented Generation (RAG) is a technique that combines information retrieval with text generation to produce more accurate and contextually relevant answers from large language models.
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
About these practice questions
One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 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-900 exam.