AI-900 Practice Question: Describe features of generative AI workloads on Azure
A marketing agency wants to use Azure OpenAI Service to generate product descriptions. They need the descriptions to be factually accurate and based on their specific product catalog, which is stored in a vector database. Which technique should they use to ground the model's outputs in their own data?
⚠ Common exam trap
It's easy for candidates to confuse fine-tuning with RAG, assuming that training the model on custom data is the only way to incorporate proprietary information, but RAG achieves the same goal more efficiently and flexibly without retraining.
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
✓
Prompt engineering with retrieval augmented generation (RAG)
Retrieval augmented generation (RAG) is the correct technique because it allows the model to retrieve relevant, up-to-date product information from the vector database at inference time and use that data as context to generate factually accurate descriptions. This grounds the model's outputs in the specific product catalog without modifying the underlying model weights, ensuring responses are based on the agency's own 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.
- ✗
Fine-tuning the model on the product catalog
Why it's wrong here
Fine-tuning modifies the model's weights using a static dataset of product descriptions, resulting in a fixed snapshot of knowledge that cannot reflect subsequent catalog updates. Because it does not involve any retrieval step at inference time, the model never sees a live vector database or current product records, so it will still produce hallucinated or outdated details for products outside its training distribution. It also requires costly per-catalog training runs and does not prevent fabrication, since generation is not grounded in retrieved evidence.
- ✓
Prompt engineering with retrieval augmented generation (RAG)
Why this is correct
Prompt engineering with retrieval augmented generation (RAG) grounds each response by embedding the user query, retrieving the most relevant product documents from a vector database, and injecting those chunks into the prompt as authoritative context. The model then generates the description while being conditioned on these retrieved facts, which dramatically reduces hallucination and ensures the text reflects the latest catalog state without retraining. Effective RAG also relies on careful prompt wording, such as instructing the model to cite only the provided context and ignore unrelated knowledge.
- ✗
Zero-shot prompting without additional data
Why it's wrong here
Zero-shot prompting provides no grounding: the model receives only the instruction and the product name, so it must generate entirely from knowledge encoded during pretraining. That parametric memory is frozen at the training cutoff, making it blind to new products, promotions, or spec changes, and it often lacks details for niche SKUs. Consequently, the output tends to be generic and can confidently describe features that do not exist for a specific item, which is unacceptable for an accurate marketing description.
- ✗
Reinforcement learning from human feedback (RLHF)
Why it's wrong here
Reinforcement learning from human feedback (RLHF) tunes the model's generation policy to produce responses that a reward model scores as helpful and safe, but it does not add any mechanism for accessing external data. The model's factual knowledge remains unchanged after RLHF, and at inference time it still cannot query a vector index or live product database. Thus, while RLHF improves conversational quality and alignment, it leaves the core problem of stale or invented product details untouched.
Go deeper
Related to this question
Learn chapter
Azure Machine Learning Studio
Key term
Service
A service is a software component or system that performs a specific function and is available to be used by other programs or users over a network.
Key term
Azure OpenAI Service
Azure OpenAI Service is a cloud platform from Microsoft that lets developers use powerful artificial intelligence models, like GPT-4, to build applications that can understand and generate human-like text, code, images, and more.
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.