AI-900 Practice Question: Describe features of generative AI workloads on Azure
A marketing team wants to use Azure OpenAI to generate blog post outlines. They have a single example of an outline that follows their preferred structure: introduction, three key points, conclusion. They want the model to generate new outlines that follow the same structure without retraining the model. Which technique should they use?
⚠ Common exam trap
Many exam-takers confuse few-shot learning with fine-tuning, assuming that any task requiring consistent output format must involve retraining the model, when in fact in-context learning via prompt engineering is sufficient for small numbers of examples.
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
✓
Providing the example outline in the prompt (few-shot learning)
Few-shot learning involves providing a small number of examples (in this case, one example outline) directly in the prompt to guide the model's output format and structure without any retraining. This technique leverages the model's in-context learning ability to mimic the given pattern, making it ideal for generating new outlines that follow the same structure.
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 a large dataset of blog outlines
Why it's wrong here
Fine-tuning the model on a large dataset of blog outlines is a supervised training process that updates the model's weights, requiring a large, labeled dataset and significant compute resources. It is a heavyweight solution for a task that only needs a one-off or generic structure, and it would permanently alter the model, potentially causing overfitting or catastrophic forgetting. Since the marketing team just wants to generate a blog post from an outline, this approach is unnecessary and impractical compared to a prompt-based method.
- ✓
Providing the example outline in the prompt (few-shot learning)
Why this is correct
Providing the example outline in the prompt leverages few-shot learning, a form of in-context learning where the model uses the examples directly from the prompt to infer the desired output format and style without any weight updates or retraining. This is the correct approach because Azure OpenAI's generation models, such as GPT-4, are designed to condition on the provided context and replicate the pattern shown in the examples. It is fast, cost-effective, and reversible, making it ideal for guiding the model to follow a specific blog outline.
- ✗
Setting the temperature parameter to a high value
Why it's wrong here
Setting the temperature parameter to a high value increases the randomness and diversity of the token sampling process, making the output more creative and unpredictable. This does not enforce or guide the model to follow a provided outline; in fact, it is more likely to cause the model to deviate from the structure and produce less coherent content. Temperature controls the probability distribution over tokens, not the semantic or structural adherence to an example, so it is the wrong tool for ensuring the output matches a specific blog outline.
- ✗
Using the Azure OpenAI embeddings API
Why it's wrong here
The Azure OpenAI embeddings API converts text inputs into high-dimensional vector representations that capture semantic meaning, which are used for tasks like measuring text similarity, clustering, or retrieval — not for generating new text. Embeddings are a non-generative capability, so calling the embeddings API would return a vector, not a blog post. To generate content, you need to use the completions or chat completions API, where the model produces text based on the prompt; embeddings alone cannot fulfill this requirement.
Go deeper
Related to this question
About these practice questions
This AI-900 question is part of Courseiva's 985-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 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.