What happened
Microsoft retired exam AI-102 and its certification, Azure AI Engineer Associate, on 30 June 2026. The free renewal assessment was retired on the same day. The replacement is exam AI-103, which leads to a new certification with a new name: Azure AI Apps and Agents Developer Associate. AI-103 is the only exam that certification requires.
The new name says what changed. AI-102 treated agents as a 5–10% topic. AI-103 is built around building agents and generative AI applications on Microsoft Foundry, in Python.
Skill areas, side by side
| AI-102 (retired) | Weight | AI-103 (current) | Weight |
|---|---|---|---|
| Plan and manage an Azure AI solution | 20–25% | Plan and manage an Azure AI solution | 25–30% |
| Implement generative AI solutions | 15–20% | Implement generative AI and agentic solutions | 30–35% |
| Implement an agentic solution | 5–10% | (merged into the line above) | |
| Implement computer vision solutions | 10–15% | Implement computer vision solutions | 10–15% |
| Implement natural language processing solutions | 15–20% | Implement text analysis solutions | 10–15% |
| Implement knowledge mining and information extraction solutions | 15–20% | Implement information extraction solutions | 10–15% |
AI-102 weights are from its last published study guide (skills as of 23 December 2025); AI-103 weights are from the current study guide (skills as of 16 April 2026). Microsoft publishes both as ranges.
What changed, in practice
- Generative AI and agents are now one area and the largest on the exam, at 30–35%: building generative applications with Foundry, building agents with Foundry, and optimising and operating them.
- Planning and managing grew to 25–30% and now explicitly includes responsible AI across generative and agentic systems, plus monitoring and securing them.
- Computer vision now covers image and video generation and multimodal understanding, not only image analysis.
- Natural language processing became text analysis with language models, plus speech.
- Knowledge mining became information extraction: retrieval and grounding pipelines, and extracting content from documents — the building blocks of retrieval-augmented generation.
If you already hold Azure AI Engineer Associate
Microsoft's credential-retirement policy says a certification earned or renewed before it was retired stays in the Active Certifications section of your transcript until it expires, then moves to Historical. Because the renewal assessment is also retired, it cannot be extended. To hold a current Azure AI credential after that date, the route is AI-103.
If you were studying for AI-102
Most of the Azure AI service knowledge carries over — vision, language, speech, document extraction, and the planning and security of AI resources. What you will need to add is depth on Microsoft Foundry: building and evaluating generative applications and agents, and grounding them with retrieval. Start from the AI-103 study guide rather than an AI-102 course, and check any practice material you use says AI-103 on it.
Sources
- Microsoft Learn — Exam AI-103 study guide
- Microsoft Learn — Exam AI-103
- Microsoft Learn — Exam AI-102 study guide (retirement notice)
- Microsoft Learn — Renewal for Azure AI Engineer Associate
- Microsoft Learn — Credential retirement
- Microsoft Tech Community — New Azure AI Apps and Agents Developer Associate