Update (September 2026): AI-900 and AI-102 retired on June 30, 2026. Their replacements, AI-901 (Azure AI Fundamentals) and AI-103 (Azure AI App and Agent Developer Associate), are the current exams. For the current path, see the AI-103 Study Guide.
The clearest breakdown of both exams, who needs which, and the smartest certification path if your goal is Azure AI development.
If you are planning your Microsoft certification path in 2026, you have probably asked this at some point: do I need AZ-900 before AI-901? Or can I skip it entirely?
The short answer is that they cover very different ground. Whether you need both depends entirely on what you are trying to build and where you want to take your career.
This guide breaks it down clearly so you can stop second-guessing and start studying the right thing.
What Is AZ-900?
AZ-900, Microsoft Azure Fundamentals, is the entry-level certification for the Azure cloud platform itself. It has nothing to do with artificial intelligence specifically. Instead it covers how Azure works as a cloud platform.
Topics covered in AZ-900 include:
- Cloud computing concepts: IaaS, PaaS, SaaS
- Core Azure services: virtual machines, storage accounts, virtual networks
- Azure identity and access: Entra ID, role-based access control
- Azure pricing, governance, and compliance
- The shared responsibility model
AZ-900 is the foundation for understanding how Azure runs as infrastructure. If you have never worked with Azure before and need to understand why networking, permissions, and storage behave the way they do, AZ-900 gives you that context.
It is also worth knowing that AZ-900 has lifetime validity. If you earn it before retirement, it never expires.
New here? Skim the free AZ-900 study guide online (11 modules, ~90 min) before you sit the exam - no email required.
What Is AI-901?
AI-901, officially titled Azure AI Fundamentals, is the 2026 replacement for the AI-900 exam. AI-900 retired on June 30, 2026 and AI-901 took its place.
But AI-901 is not just a renamed version of AI-900. It has been significantly updated to reflect where Azure AI actually is in 2026.
Topics covered in AI-901 include:
- Core AI and machine learning concepts
- Responsible AI principles
- Generative AI concepts and large language models
- Azure OpenAI Service capabilities
- Introduction to Microsoft Foundry
- Building basic AI apps and agents on Azure
- Prompt engineering fundamentals
- AI workload identification and service selection
The key shift from AI-900 to AI-901 is the move from understanding AI services to understanding how to build with them. AI-901 introduces Microsoft Foundry and AI agents at a conceptual level, which AI-900 barely touched.
Try a real Azure AI lab, free
A pre-provisioned Azure environment in your browser. No Azure account, no credit card, no setup.
Most large enterprises already run on Microsoft infrastructure. Azure AI Foundry is the platform those teams use to build and deploy AI apps, so the skills you practice here map directly to real job demand.
AZ-900 vs AI-901: Side by Side
The main difference between AZ-900 and AI-901 is that AZ-900 teaches the core infrastructure of the Azure cloud, including networking, storage, and security, while AI-901 focuses exclusively on artificial intelligence concepts, large language models, and Microsoft Foundry.
| Feature | AZ-900 | AI-901 |
|---|---|---|
| Focus | Azure cloud infrastructure | Azure AI fundamentals |
| Level | Fundamentals | Fundamentals |
| Coding required | No | No |
| Azure OpenAI | Not covered | Core topic |
| Microsoft Foundry | Not covered | Introduced |
| AI agents | Not covered | Conceptual coverage |
| Networking and storage | Core topic | Not covered |
| Entra ID and permissions | Core topic | Not covered |
| Responsible AI | Not covered | Core topic |
| Who it is for | Anyone starting with Azure | Anyone starting with Azure AI |
| Status | Current | Current, replaced AI-900 in June 2026 |
Do You Need Both?
It depends entirely on your goal.
If you are a developer building AI applications on Azure
You do not need to pay for AZ-900. However, you should understand the infrastructure basics because you will hit problems with networking, permissions, and storage when connecting AI services to real company data. A free AZ-900 crash course on YouTube covers what you need without the exam cost.
Your actual exam effort should go into AI-901 and then AI-103.
If you are completely new to Azure
AZ-900 first makes sense. It gives you the cloud foundation and then AI-901 builds the AI layer on top. The two exams complement each other with minimal overlap.
If your company is migrating to Azure and you need both cloud and AI skills
Both are worth doing. You will need AZ-900 level knowledge to understand how Azure infrastructure works before you can confidently connect AI models to company systems, private data, and internal networks.
If you are a non-technical professional who needs AI literacy
AI-901 is the right choice. Skip AZ-900 entirely. AI-901 does not require technical experience and gives you exactly what you need to understand and communicate about AI initiatives in your organisation.
The 2026 Microsoft AI Certification Path
If you are serious about building a career in Azure AI development, here is the recommended path:
Step 1: AI-901
Get the AI fundamentals in place. Understand what Azure AI services exist, when to use them, and how Microsoft Foundry fits in. Think of this as learning the landscape before you start building.
Step 2: AI-103
This is where things get real. AI-103, the Azure AI App and Agent Developer Associate certification, is built for developers who want to actually build AI applications and agents in production. You will go hands-on with Azure AI Foundry, Azure OpenAI, RAG patterns, agent orchestration, prompt engineering, and responsible AI implementation. This is the certification that matters most for Azure AI developers in 2026.
Step 3: AI-300 (optional)
If you want to go into MLOps, building production pipelines, and operationalising machine learning at scale, AI-300 is the next step after AI-103.
Still deciding? Grab the AI-901 exam study guide
One-page reference covering Azure AI Fundamentals topics, Microsoft Foundry, and exam tips.
What About Career Switchers?
If you are switching into cloud or AI from a different background, both exams are genuinely useful. AZ-900 proves you understand how enterprise cloud infrastructure works. AI-901 proves you understand how AI is built and deployed on that infrastructure.
Together they give you a credible foundation to enter either a cloud engineering or AI development role. Neither is a hard prerequisite for the other, but starting with AZ-900 if you have no cloud experience at all is the sensible approach.
The Bottom Line
AZ-900 and AI-901 are not competing exams. They cover different domains and serve different purposes.
If your goal is Azure AI certification in 2026, AI-901 is your starting point. You do not need AZ-900 to take it and you do not need AZ-900 to take AI-103 after it.
But if you want to build real AI systems on Azure that connect to company infrastructure, private networks, and secure data stores, understanding the AZ-900 content will save you a lot of debugging time. Whether you pay for the exam or just go through free materials is your call.
The most important thing is not to spend months on AZ-900 when AI-901 and AI-103 are the certifications that will actually move your career forward in 2026.
Prepare for AI-901 and AI-103 with Hands-On Labs
At GetHandsOn.ai we build hands-on lab environments specifically for the new 2026 Microsoft AI certifications. Our AI-901 and AI-103 bundles include real Azure lab environments, practice questions mapped to exam objectives, and condensed study guides that cut your prep time without cutting corners.
Both the AI-901 and AI-103 lab courses are live today. One-time payment, three months of access.
Explore the AI-901 and AI-103 bundles
If you only have time for one
Most people asking this question are not really choosing between two exams, they are asking which one to skip. The honest answer depends on the job you want next.
You want to work on AI features. Take AI-901 and skip AZ-900. Everything AZ-900 covers that you will actually need on an AI project, mainly resource groups, regions, identity and cost, gets covered again in context when you build something. AZ-900 on its own does not put you closer to an AI role.
You work in IT operations or want a broad cloud grounding. Take AZ-900 first. It is a better map of Azure as a whole, and AI-901 will make more sense afterwards because you will already know what a resource group and a managed identity are.
You are already a developer. Consider skipping both and going to AI-103. It is the associate-level exam and it is what hiring managers screen for on AI roles. AI-901 is useful mainly as a confidence check before committing to a harder exam.
The one combination we would question is taking AZ-900 and AI-901 back to back purely for the credential count. Two fundamentals certifications signal the same thing to a recruiter as one, and the time is better spent on hands-on work or on the associate exam.