Browse Microsoft Certification Guides

Azure AI-900 Generative AI Guide

Study Azure AI-900 Generative AI: key concepts, common traps, and exam decision cues.

This chapter covers the most recent part of the blueprint and also the most likely source of overthinking. AI-900 does not expect advanced prompt engineering or agent design. It expects you to understand when generative AI is the right workload, why grounding matters, and what Microsoft’s Azure generative-AI platform surfaces do at a high level.

Current weight in the study guide

Microsoft currently weights this skill area at 20-25% of the exam.

Work this skill area in order

Lesson Focus
5.1 Generative AI Scenarios, Grounding and Responsible Use Learn when the problem is truly generative and how grounding changes answer quality.
5.2 Azure AI Foundry, Azure OpenAI and Model Catalog Basics Learn the Azure generative-AI platform boundaries AI-900 expects.

Fast routing inside this chapter

If the question is really about… Go first to…
deciding whether generation is appropriate 5.1 Generative AI Scenarios, Grounding and Responsible Use
choosing the Azure generative-AI platform surface 5.2 Azure AI Foundry, Azure OpenAI and Model Catalog Basics

What strong answers usually do

  • choose generative AI only when the task truly needs new content
  • prefer grounding when the answer must stay close to trusted internal information
  • keep responsible-AI considerations attached to generative use cases, not separate from them
  • recognize the Microsoft Foundry rename without assuming the objective boundary changed

In this section

Revised on Sunday, May 10, 2026