Courseiva

AI-900 Practice Question: Describe features of generative AI workloads on Azure

What is the primary benefit of using Retrieval Augmented Generation (RAG) over relying solely on an LLM's trained knowledge?

⚠ Common exam trap

Test-takers frequently confuse RAG with general LLM optimization techniques (like fine-tuning or prompt engineering) and assume it improves speed or reduces cost, when in fact its primary value is factual grounding and recency.

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

RAG grounds LLM responses in current, specific information — reducing hallucination and knowledge cutoff issues

RAG enhances LLM outputs by retrieving relevant, up-to-date information from an external knowledge base (e.g., Azure Cognitive Search) and injecting it into the prompt context. This grounds the model's response in verifiable data, significantly reducing hallucinations and overcoming the knowledge cutoff limitation inherent in static training data.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

  • RAG makes LLMs faster by skipping the training process

    Why it's wrong here

    RAG is a framework for inference, not a way to avoid training; the underlying LLM is still trained, and the retrieval system has its own embeddings/index. At query time, relevant documents are retrieved and inserted into the prompt, so the model's weights are frozen and no additional training or fine-tuning occurs. While RAG can help adapt to new information without retraining, it does not make the model faster — in fact, retrieval adds overhead and can increase response latency.

  • RAG grounds LLM responses in current, specific information — reducing hallucination and knowledge cutoff issues

    Why this is correct

    RAG bridges the gap between static training data and dynamic world knowledge. Instead of asking the LLM to recall facts from memory, the system first runs a similarity search against a vector index of documents, then conditions the generation on the retrieved passages. This changes the model's behavior from memorization to evidence-based reasoning, dramatically lowering the likelihood of fabricated or outdated output.

  • RAG reduces the cost of API calls by batching requests

    Why it's wrong here

    RAG adds a retrieval step before generation, which often introduces additional API calls or infrastructure (e.g., vector database queries) and injects extra tokens into the prompt, so it can increase rather than decrease cost. Batching, by contrast, is a throughput optimization that groups multiple requests to reduce per-request overhead. Cost reduction is not a design goal of RAG; its value lies in answer quality and freshness.

  • RAG allows LLMs to process images alongside text

    Why it's wrong here

    Retrieval-Augmented Generation retrieves relevant text passages from an external knowledge source and inserts them into the LLM's context window; it has no mechanism for encoding or interpreting image content. Multimodal tasks like image captioning or visual question answering require specialized multimodal models (e.g., vision-language models). Thus RAG does not enable image processing.

About these practice questions

Courseiva writes every AI-900 question from scratch — 985 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 →

How Courseiva writes practice questions · Editorial policy

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.