The Azure AI Fundamentals AI-900 exam is retiring on 30th June and is being replaced by the AI-901 exam. There is quite a change in emphasis between the exams, so I thought I would explain the differences.
Before we start, the certification is not retiring, just the exam required to earn it, so if you have passed AI-900 then you don’t need to take AI-901.
AI-900
AI-900 is about artificial intelligence concepts, machine learning concepts, Machine Learning in Azure, and prebuilt class AI services around language, vision, and information extraction.
Course
The course has the following major themes:
- AI
- Machine learning
- Natural language processing
- Computer vision
- AI-powered information extraction
Each of these topics has two modules. The first is high-level concepts and the second covers the services provided by Microsoft Azure.
The focus is on the classic Azure ML and AI Services, formerly known as Cognitive Services, that has been available for many years. Recently the course added an introduction to Microsoft Foundry.
Exam
The AI-900 exam tests knowledge of machine learning and AI concepts and related Azure services.
The exam covers five areas:
- Describe Artificial Intelligence workloads and considerations
- Describe fundamental principles of machine learning on Azure
- Describe features of computer vision workloads on Azure
- Describe features of Natural Language Processing (NLP) workloads on Azure
- Describe features of generative AI workloads on Azure
Each of these is split between concepts and the features and capabilities of the services provided by Azure. Candidates need to be able to distinguish between the different services for a scenario.
AI-901
AI-901 is about artificial intelligence concepts and building AI-powered apps using Microsoft Foundry.
Course
The course has the following major themes:
- AI
- Generative AI and agents
- Natural language processing
- Speech
- Computer vision
- Generative AI-powered information extraction
Each of these topics has two modules. The first is high-level concepts and the second covers the services provided by Microsoft Azure to support those concepts.
The focus is on generative AI services and tools in Microsoft Foundry. You don’t need to have any experience of using Microsoft Foundry before taking this course, but a basic familiarity with coding (Python preferred) is assumed.
Exam
The AI-901 exam tests knowledge of AI solutions in Microsoft Foundry and the foundational technical skills to work with them.
The AI-901 exam requires knowledge of Python coding syntax, and programming techniques, and SDKs.
The exam covers two areas:
- Identify AI concepts and responsibilities
- Implement AI solutions by using Microsoft Foundry
The first of these is about AI concepts, generative AI models, and AI workloads. Candidates need to be able to distinguish between the features and capabilities of these workloads.
The second area is about the features and capabilities of generative AI, agents, natural language, vision, and information extract services and tools provided in Foundry. Candidates need to be able to build apps using these services.
Exam comparison
|
Area |
AI‑900 |
AI‑901 |
|
Audience |
Nontechnical or technical beginners |
Technical beginners planning to build AI solutions |
|
Skills tested |
Understanding AI and machine learning concepts; identifying when to use Azure AI services |
Understanding AI concepts; implementing AI solutions with Microsoft Foundry |
|
Coding knowledge |
None |
Basic Python syntax and programming concepts |
|
Knowledge level |
Introductory, conceptual understanding of Azure AI services (classic approach) |
Foundational understanding of Azure resources, plus implementation with Foundry, provisioning Azure resources, and modern AI development tooling |
|
Focus |
What is AI? |
How do I build an AI app by using Foundry? |