A professional services firm, Contoso Advisory, wants to build an internal “Ask the Firm” assistant that answers questions over proposals…

The statement is true. It accurately describes the Retrieval-Augmented Generation (RAG) pattern, which is a standard and highly effective architecture for building enterprise question-answering systems. In this pattern, Azure AI Search serves as the specialized retrieval system. It indexes the firm's private documents and uses its advanced search capabilities (full-text, semantic, and vector) to find the most relevant information for a user's query. This retrieved content is then passed as context to an Azure OpenAI model, which "grounds" its response in the provided facts, reducing hallucinations and improving accuracy.

This is a true/false question, so there are no other options to evaluate. The provided statement accurately reflects a primary and officially documented use case for these Azure services.

1. Microsoft Learn. (2024). Retrieval Augmented Generation (RAG) in Azure AI Search. "Azure AI Search is a foundational component in RAG architectures... It offers a range of search capabilities, including keyword, semantic, and vector search, to retrieve relevant data that can be used by a generative AI model."

2. Azure Architecture Center. (2024). Build an enterprise chat solution with Azure OpenAI. This reference architecture explicitly shows Azure AI Search providing the indexed data and retrieval function to ground a large language model from the Azure OpenAI Service.