AI-900 Practice Question: Describe features of generative AI workloads on Azure
A developer wants to use Azure OpenAI to generate text that follows a specific style, such as formal business letters. They provide three examples of the desired output format in the prompt and then ask the model to generate a new letter. Which technique is the developer using?
⚠ Common exam trap
Watch out — candidates often confuse few-shot learning with fine-tuning, mistakenly thinking that providing examples in a prompt is equivalent to training the model, when in fact fine-tuning involves updating model parameters through additional training on a dataset.
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
✓
Few-shot learning
The developer is using few-shot learning, a technique where a prompt includes several examples (in this case, three formal business letters) to guide the model's output style and format without updating the model's weights. This approach leverages the model's in-context learning ability to generalize from the provided examples, making it ideal for tasks requiring specific stylistic adherence.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Zero-shot learning
Why it's wrong here
Zero-shot learning means the prompt contains no illustrative input-output pairs, so Azure OpenAI must infer the desired format entirely from its prior training. While this can handle common, well-represented patterns, it often produces inconsistent structure for a specific user-defined format because no in-context demonstration constrains the output. Here, zero-shot learning would not reliably enforce the exact pattern requested.
- ✓
Few-shot learning
Why this is correct
Few-shot learning supplies two or more complete input-output examples inside the prompt, letting Azure OpenAI infer the pattern through in-context learning without modifying model weights. The examples act as implicit constraints that shape the generated text to match the specified format on new inputs. This is the correct approach when a specific format must be followed and a handful of demonstrations are available.
- ✗
Fine-tuning
Why it's wrong here
Fine-tuning updates the model's weights by retraining on a large, labeled dataset through a supervised learning process, creating a custom model deployment. It is a separate, compute-intensive step done before inference, not a runtime technique for steering a single prompt. Adding a few examples to a prompt is far cheaper and does not change model parameters, so fine-tuning is not the right mechanism here.
- ✗
Temperature scaling
Why it's wrong here
Temperature scaling is a sampling parameter that adjusts the probability distribution over predicted tokens, making outputs more deterministic at low values and more random at higher values. It affects vocabulary choice and creativity but provides no information about the desired output structure or format. Thus it cannot teach the model to follow a specific pattern; it only controls variability of responses.
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.