AI-901 Microsoft Azure AI Fundamentals

What's inside

AI-901 doesn't test buzzwords. It tests whether you know which AI workload fits a scenario, which setting shapes a model's answer, and which responsible AI principle is at stake.

This independent study guide for the Microsoft Certified: Azure AI Fundamentals exam distils what AI-901 really expects you to understand into clear explanations, comparisons and the traps that catch beginners. AI-901 replaced AI-900 in 2026, so the book carries forward the AI-900 knowledge that is still examined, in today's product names, and adds the hands-on Microsoft Foundry skills the new exam asks about.

Organised by technology

Nine chapters, each readable on its own and together covering both AI-901 skills areas:

• AI concepts: the six responsible AI principles and the four mitigation layers, how generative models turn prompts into answers, choosing a model by capability, temperature, Top P and max tokens, and recognising AI workload types.
• Microsoft Foundry: the model catalog and deployments, calling a deployed model from code with keyless authentication, system and user prompts, and building and calling agents with Foundry Agent Service.
• Language, speech and vision: text analysis and translation, Azure Speech and Voice Live, images in prompts, Azure Vision and image generation with gpt-image models.
• Information extraction: Azure Content Understanding analyzers for documents, images, audio and video.

Inside

• Comparison tables for every close call, such as which principle, which workload or which model setting fits a requirement.
• 30+ Common trap callouts that name the answers that look right but aren't, and explain what's right instead.
• 7 diagrams, from the mitigation layers and next-token prediction to agent anatomy and Content Understanding.
• Key takeaways at the end of every chapter for fast revision.
• A glossary of 200+ terms, each linked to Microsoft Learn, with a free online version for print readers.

Up to date

Azure AI changes quickly. This edition reflects Microsoft's documentation as of October 2026, including recent retirements and renamed products, and notes where older material and today's product differ.

Understanding, not memorising

This book contains no exam questions. It explains the knowledge the exam expects, so you can answer questions you have never seen and talk confidently about AI solutions at work.

Written by Tony Rough, a cloud architect with more than twenty years in IT infrastructure who holds the Azure Solutions Architect Expert, Azure Administrator and Azure Security Engineer certifications.

Part of the Ultra Transcenders series from Distilled Press. An independent publication, not affiliated with, sponsored by or endorsed by Microsoft Corporation.

Chapters

1. Responsible AI
2. How generative AI models work and how to choose one
3. AI workloads
4. Microsoft Foundry: models, deployments and code
5. Prompts and agents
6. Text analysis and translation
7. Speech: recognition, synthesis and voice agents
8. Computer vision and image generation
9. Information extraction with Azure Content Understanding

Plus an appendix glossary of 200+ terms, each linked to Microsoft Learn, and free online for print readers.

Sample diagrams

Free study notes

Some sections of the book, free to read online:

• The six Responsible AI principles, with examples

  Fairness, reliability and safety, privacy and security, inclusiveness, transparency and accountability, and how to tell them apart in a scenario.

• Responsible generative AI: guardrails and the four mitigation layers

  How guardrails, system messages, grounding and user experience design reduce harm in a generative AI solution.

• How generative AI models work: tokens, attention and next-token prediction

  What happens between a prompt and a response, and how inference differs from training.

• Temperature, Top P and max completion tokens explained

  Which model setting controls randomness, which controls length and cost, and which ones are not set at deployment.

• AI workload types: generative, agentic, text, speech, vision and extraction

  How to recognise each AI workload from a scenario.

• What an AI agent is, and how tool_choice controls tool calls

  Agents as model plus instructions, knowledge and tools, and the auto, required and none tool_choice values.

• Azure Speech capabilities at a glance

  Speech to text, text to speech, translation, batch transcription and speaker recognition compared.

• Azure Content Understanding: analyzers and structured extraction

  How analyzers turn documents, images, audio and video into structured JSON, and how to call them from code.