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

Answer analysis

Option-by-option breakdown

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

• ✗

  Using Bing Maps to provide location-based responses

  Why it's wrong here

  Bing Maps supplies spatial data such as geographic coordinates, routes, and map tiles, which power location-based responses; it is not a general-purpose web search grounding mechanism. Bing search grounding instead queries the Bing web index to retrieve up-to-date natural-language or web content relevant to a prompt. While a grounder could surface a map as a search result, the service itself is about augmenting the LLM with current text-based web information, not about providing mapping or geocoding functions.

• ✓

  Retrieving current web information from Bing to augment LLM responses beyond its training cutoff

  Why this is correct

  Bing search grounding works by sending the model's prompt or a generated search query to the Bing Search API at inference time, retrieving current web pages and snippets, and inserting the relevant content into the model's context before it composes an answer. This lets the model reference events, facts, and data that post-date its training cutoff, reducing hallucination and improving factual accuracy. The retrieved information is used as additional context, not encoded into the model's weights, so it remains a retrieval-augmentation technique rather than fine-tuning.

• ✗

  Translating Copilot responses using Microsoft's Bing Translator

  Why it's wrong here

  Translating Copilot responses into another language is a text-transformation operation performed by Azure AI Translator, a dedicated machine translation service; it does not depend on web search or on grounding the model with current documents. Bing search grounding, by contrast, is an information-retrieval mechanism that supplies the LLM with freshly fetched web content to ground its answers in verifiable, up-to-date sources. Confusing the two conflates the final vocabulary-adaptation step with the retrieval of facts needed to answer the question correctly.

• ✗

  Using Bing advertising data to personalize AI responses

  Why it's wrong here

  Bing advertising data describes user profiles, demographics, and click-through patterns used for ad targeting, and leveraging it to personalize AI responses would be a completely different, user-modeling feature. Bing search grounding does not track an individual's preferences to tailor answers; it simply takes the current user prompt and retrieves relevant, publicly available web pages. The goal is factual grounding in the live web index, not behavioral personalization, so ad data is unrelated to how grounded responses are generated.

Go deeper

Related to this question