- What Does "AI-102" Actually Mean?
- Why AI-102 Now Points to a Retired Exam
- What AI-102 Means for People Who Already Hold It
- What the Successor Credential Covers
- Registration, Fees, and Scoring Mechanics
- Who This Certification Is Meant For
- Turning the Meaning Into a Study Plan
- Frequently Asked Questions
- AI-102 is the exam code for a Microsoft credential that retired on June 30, 2026, at 11:59 PM Central.
- The live successor is AI-103, Developing AI Apps and Agents on Azure, awarding the Azure AI Apps and Agents Developer Associate title.
- Implement generative AI and agentic solutions is the heaviest domain today, at 30-35% of the exam.
- Exam cost is $165 USD at standard US pricing, with a 700/1000 passing score over a 120-minute session.
What Does "AI-102" Actually Mean?
"AI-102" is a Microsoft exam code, not a job title or a skill in itself. It refers to the exam formally named Designing and Implementing a Microsoft Azure AI Solution, which awarded the credential Microsoft Certified: Azure AI Engineer Associate. The number "102" follows Microsoft's internal numbering convention for its role-based certification exams, and for years this code was the primary way developers signaled that they could build production AI solutions on Azure.
Here is the part that trips people up in 2026: the exam code AI-102 and the credential it represented are no longer live. Microsoft retired both the exam and the associated credential, along with its renewal assessment, on June 30, 2026, at 11:59 PM Central. When people search "AI-102 meaning" today, they are usually trying to understand one of two things - either what the term used to represent, or what replaced it. This article answers both, and if you want the full history and status breakdown, the What Is AI-102? guide covers that in more depth.
Why AI-102 Now Points to a Retired Exam
Microsoft periodically retires exams when the underlying technology shifts enough that the old blueprint no longer reflects what practitioners actually build. That's exactly what happened here. Azure's AI stack moved decisively toward agentic patterns - orchestrated multi-agent systems, tool-calling, and managed agent runtimes - and the AI-102 blueprint, built around discrete cognitive services, could not keep pace. Microsoft responded by retiring AI-102 entirely rather than patching it, and launching AI-103, Developing AI Apps and Agents on Azure, as the direct successor.
At the training level, the four-day AI-103T00-A course has replaced the older AI-102T00 course, reinforcing that this is a full succession, not a side-by-side option. If your goal is to actually earn a certification today, the only path forward under this exam family is AI-103. For a broader breakdown of what "AI-102" is understood to mean across its full lifecycle - including how the term is used loosely to refer to the whole certification lineage - see the companion piece on AI-102 meaning and what does AI-102 mean for additional framing.
Key Takeaway
If you're planning your certification path from scratch in 2026, don't study for AI-102. Study for AI-103. The content, exam code, and credential name have all changed, even though the subject area - building AI solutions on Azure - is a direct continuation.
What AI-102 Means for People Who Already Hold It
If you passed AI-102 before its retirement date, the credential doesn't disappear from your record. It remains listed in the Active Certifications section of your Microsoft transcript until its printed expiration date. What changes is the path forward: there is no renewal assessment available anymore, because the renewal assessment retired alongside the exam itself. Once your certification's printed expiration passes, there is no mechanism to renew it under the old code - the only route to staying current in this domain is to pursue AI-103 directly.
This matters for how you talk about your credentials on a resume or LinkedIn profile. The credential is technically accurate to list while active, but hiring managers scanning job postings in 2026 and beyond will increasingly look for AI-103, since that's the certification tied to current, generally-available Azure AI tooling. For a deeper look at how this status change affects long-term career value, see Is the AI-102 Certification Worth It?
| Attribute | AI-102 (Retired) | AI-103 (Current) |
|---|---|---|
| Exam Title | Designing and Implementing a Microsoft Azure AI Solution | Developing AI Apps and Agents on Azure |
| Credential Name | Azure AI Engineer Associate | Azure AI Apps and Agents Developer Associate |
| Status | Retired June 30, 2026, 11:59 PM Central | Live and earnable |
| Associated Training | AI-102T00 | AI-103T00-A (four-day course) |
| Renewal | No renewal path available | Free annual renewal via Microsoft Learn assessment |
What the Successor Credential Covers
Since AI-103 is the exam that actually determines what "AI-102 knowledge" means in practice today, understanding its five domains is essential. The weighting tells you exactly where to invest your study hours. For a complete section-by-section breakdown, read the AI-102 Exam Domains 2026: Complete Guide to All 5 Content Areas.
Domain 1: Plan and manage an Azure AI solution (25-30%)
This domain tests whether you can architect and govern an AI solution before writing a line of application code.
- Selecting the right Azure AI services and Foundry resources for a given scenario
- Security, cost, and responsible AI planning
- Resource provisioning and lifecycle management
Domain 2: Implement generative AI and agentic solutions (30-35%)
This is the heaviest domain on the exam and the clearest signal of how far the field has moved since the original AI-102. Expect deep coverage of Microsoft Foundry, Foundry Agent Service, and the Microsoft Agent Framework.
- RAG implementation patterns and grounding data
- Multi-agent orchestration and autonomous workflows with approval controls
- Function calling, tool schemas, and conversation memory
- Agent evaluation, error analysis, tracing, and token analytics
Domain 3: Implement computer vision solutions (10-15%)
Covers image analysis, spatial analysis, and vision-related Foundry Tools integrations.
Domain 4: Implement text analysis solutions (10-15%)
Covers natural language processing tasks, including sentiment, key phrase extraction, and language detection using Azure AI services.
Domain 5: Implement information extraction solutions (10-15%)
Focuses on structured extraction from documents and content, including Azure Content Understanding in Foundry Tools and Azure Document Intelligence in Foundry Tools.
Notice how much of the blueprint now assumes fluency with agent-oriented services - Foundry Agent Service, the Agent Framework, Azure OpenAI in Foundry Models - alongside the more traditional Azure AI Search, Azure Speech, and Azure Translator skill areas. If you're unsure how demanding this mix actually is compared to other associate-level exams, How Hard Is the AI-102 Exam? Complete Difficulty Guide 2026 walks through the difficulty profile in detail.
Registration, Fees, and Scoring Mechanics
Understanding what AI-102 "means" also means understanding the mechanics of actually sitting the exam under its current form. These details apply to the live AI-103 exam that carries this certification forward:
- Delivery: Owned by Microsoft, delivered through Pearson VUE, at a test center or via online proctoring.
- Price: $165 USD at standard US pricing; this varies by the country or region where the exam is proctored.
- Duration: 120 minutes, proctored, and may include interactive components.
- Passing score: 700 on a 100-1000 scale. Microsoft does not publish a fixed item count or pass rate.
- Retakes: A failed first attempt can be retaken after 24 hours, with progressively longer waits for subsequent attempts.
- Renewal: Associate-level certifications expire annually and renew free through an unproctored online assessment on Microsoft Learn.
Skills measured are current as of April 16, 2026, and most questions cover generally available features, though commonly used preview features may appear. For a complete pricing walkthrough including regional variation, see AI-102 Certification Cost 2026: Complete Pricing Breakdown, and for the exact scoring mechanics, check AI-102 Passing Score 2026: Exactly What You Need to Pass. If timing your attempt around retirement dates or testing windows matters to your plan, review AI-102 Exam Dates 2026: Testing Windows, Deadlines & Scheduling.
Who This Certification Is Meant For
The audience profile behind this exam code has always been the same type of professional: a developer, not a data scientist, who builds applications that consume AI services rather than train models from scratch. That hasn't changed with the shift to AI-103 - if anything, it's sharper now, since the heaviest domain (generative AI and agentic solutions) is squarely an application-development skill set: wiring up tool calls, managing conversation state, and building approval-gated autonomous workflows.
In practice, this credential tends to matter to hiring managers looking for people who can:
- Integrate Azure OpenAI in Foundry Models and Foundry Agent Service into production applications
- Design retrieval-augmented generation pipelines using Azure AI Search
- Build multi-agent systems with the Microsoft Agent Framework, including error handling and tracing
- Extract structured information from documents using Foundry Tools like Content Understanding and Document Intelligence
- Layer in speech and translation capabilities using Azure Speech and Azure Translator
If you're weighing whether this is worth pursuing for your career trajectory, AI-102 Salary Guide 2026: Complete Earnings Analysis and AI-102 Jobs both dig into how employers value this skill set in job postings and compensation bands. And if you want a plain-language primer before committing study time, What Is AI-102 Certification? is a good starting point.
Turning the Meaning Into a Study Plan
Once you understand what the certification actually measures, the study plan almost writes itself around domain weight. Because Implement generative AI and agentic solutions carries 30-35% of the exam, it deserves the largest and earliest block of dedicated study time - not an afterthought squeezed in near test day.
Foundation: Plan and manage an Azure AI solution
- Map out Foundry resource types, security models, and responsible AI planning before touching agent code
Deep Focus: Generative AI and agentic solutions
- Build hands-on RAG pipelines, multi-agent orchestrations, and tool-calling schemas since this domain carries the most exam weight
Vision and Text
- Work through computer vision and text analysis scenarios back to back since both sit at 10-15%
Information Extraction and Review
- Practice document extraction with Content Understanding and Document Intelligence, then run full-length practice sessions
For a complete week-by-week study framework with resource recommendations, see the AI-102 Study Guide 2026: How to Pass on Your First Attempt. Pairing that plan with realistic practice questions on our practice test platform is one of the most reliable ways to see where your agentic-solutions knowledge has gaps before exam day. You can also keep a condensed reference nearby using the AI-102 Cheat Sheet 2026: One-Page Review of Must-Know Facts during your final review week.
Key Takeaway
Don't split study time evenly across five domains. Give generative AI and agentic solutions roughly a third of your total prep time to match its 30-35% weight on the exam.
Frequently Asked Questions
Yes, historically. AI-102 was the exam code for Designing and Implementing a Microsoft Azure AI Solution and its Azure AI Engineer Associate credential. It retired on June 30, 2026, at 11:59 PM Central, but it remains a meaningful reference point for anyone who earned it before that date, and its subject area lives on through AI-103.
No. Both the exam and its renewal assessment retired and can no longer be scheduled or earned. Anyone pursuing a new certification in this lineage must take AI-103, Developing AI Apps and Agents on Azure.
AI-103, which awards Microsoft Certified: Azure AI Apps and Agents Developer Associate. The associated training also updated, with AI-103T00-A replacing the older AI-102T00 course.
No. It stays listed under Active Certifications on your Microsoft transcript until its printed expiration date. There is simply no renewal path available once that date passes, since the renewal assessment retired alongside the exam.
Partially. Domains like computer vision, text analysis, and information extraction carry forward conceptually, but the heaviest domain now centers on generative AI and agentic solutions - covering RAG, multi-agent orchestration, and tool calling - which reflects a substantial shift from the original blueprint.