AI-900 Practice Question: Describe features of generative AI workloads on Azure
A marketing team wants to use Azure OpenAI to generate blog posts. They require the output to avoid toxic language and adhere to their brand safety guidelines. Which Azure OpenAI feature should they configure to automatically block harmful content?
⚠ Common exam trap
It's easy for candidates to confuse content filters with other prompt engineering techniques like grounding or few-shot learning, assuming those can enforce safety rules, but only content filters provide automated, policy-based blocking of harmful language.
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
✓
Content filters
Content filters in Azure OpenAI are designed to automatically detect and block harmful content, including toxic language, hate speech, and violence, based on configurable severity levels. This feature directly addresses the marketing team's requirement to enforce brand safety guidelines by filtering out undesirable outputs before they are returned to the user.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Content filters
Why this is correct
Azure OpenAI's content filters are a dedicated safety layer that assesses both prompt and completion text against four category-specific models — hate, self-harm, sexual, and violence — and assigns severity levels (safe, low, medium, high). When generating a blog, these filters actively block or annotate outputs that exceed a configurable severity threshold, preventing harmful or toxic language from appearing regardless of the prompt. This is the intended mechanism to enforce content policy and is the correct answer.
- ✗
Grounding
Why it's wrong here
Grounding is a technique that anchors a model's responses to a specified data source, for example via retrieval-augmented generation, so that answers are based on external documents or databases rather than the model's broad internal knowledge. While this improves factual accuracy and reduces hallucinations, it performs no semantic safety classification — a grounded response pulled from a source could still contain hateful or violent statements if that source itself is offensive. Grounding addresses correctness, not content safety, so it would not block harmful blog text.
- ✗
Temperature
Why it's wrong here
Temperature is an inference hyperparameter that controls the randomness of token sampling: values near 0 produce more deterministic, conservative outputs, while higher values increase diversity and creativity. It shapes the flow, vocabulary, and unpredictability of generated prose, but it has no knowledge of safety categories and cannot detect or block content like hate speech or self-harm. Changing temperature merely adjusts stylistic randomness; it does not invoke any toxicity filter, so it cannot ensure a blog meets safety policies.
- ✗
Few-shot learning
Why it's wrong here
Few-shot learning supplies the model with a small set of example prompt-completion pairs to condition the format, tone, or structural patterns of its response. For a blog, examples might teach a consistent voice or header style, but few-shot examples are just part of the prompt — they do not activate a separate safety mechanism. The model could still produce a harmful completion even if the examples are benign, because the examples do not explicitly constrain the model's underlying content filters. It is a prompting technique for steered output, not a policy-enforcement system.
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.