AI-103 Practice Question 75: Provisioned Deployment Retirement

Microsoft Azure AI Apps and Agents Developer Associate (AI-103) · Test 8 · question 5 of 10

Implement generative AI and agentic solutionsmedium

A team runs a customer-facing chatbot on a Global Provisioned (PTU) deployment of a specific GPT model version. Six weeks ago they received the model's retirement-date notice by email and in Azure Service Health, but the backlog got deprioritized and no one migrated the deployment. The retirement date has now passed.

What is the state of the chatbot's deployment today?

Show answer and explanation

Answer:

D. It stopped serving inference entirely; requests now fail until the team manually migrates the deployment to a supported model

Microsoft's lifecycle policy is explicit that provisioned deployments are not auto-upgraded: "provisioned customers must manually migrate to the replacement model." That is precisely why active notification exists: a GA model's retirement notice goes out at least 60 days ahead so a provisioned customer has time to run an in-place or side-by-side migration before the cutover. Missing that window doesn't pause the clock; once retired, a model version returns error responses for every inference call, and retirement dates aren't extendable.

- A is wrong because automatic upgrade behavior (OnceNewDefaultVersionAvailable / OnceCurrentVersionExpired) applies to Standard deployment types, not Provisioned. The team cannot rely on Standard's safety net here. - B is wrong because spillover is a feature you configure in advance on a provisioned deployment to redirect traffic when PTUs are *saturated* by load; it isn't triggered by the model version itself being retired, and it requires the standard target to be named ahead of time. - C is wrong because Microsoft states plainly that retirement dates aren't extendable; missed notices don't earn a grace period.

Reference: Foundry Models lifecycle and support policy

Practice Test 8 ›

Study guides for AI-103

AI engineer roadmap 2026: skills, certs and the orderAI engineer roadmap for 2026: skills to learn in order, AI engineer vs ML engineer, which certifications fit each stage, and what to skip.

AI-200 exam guide: Azure AI Cloud Developer AssociateAI-200 replaced AZ-204: what the Azure AI Cloud Developer exam tests, how it differs from AI-103, price in USD and INR, format and how to prepare.

How AI Is Changing Software Jobs: Where to PrepareHow AI is changing software jobs: what history shows, the evidence on junior developers, what Klarna learned, new AI roles and five domains to prepare for.

AI-901 exam guide: Azure AI Fundamentals after AI-900AI-901 replaced AI-900: learn what changed, how much Python and Microsoft Foundry you need, exam costs in India, and how to prepare.

Best AI certifications 2026: AWS, Google, Microsoft, ClaudeThe best AI certifications in 2026 from AWS, Google, Microsoft, Claude and Databricks, by level, with price, status and which one to take first.

Microsoft AI certification path 2026: which exam firstMicrosoft AI certification path for 2026 - AI-901, AB-730, AI-103, AI-200, AI-300, AB-100, AI-500 and GH-300 by role, and which exam to take first.

AI-103 Certification Guide: What Changed From AI-102The AI-103 certification replaced AI-102 in 2026. Exam format, cost, the five skill domains, what changed and how to prepare, with free practice questions.

What Is RAG? How Retrieval-Augmented Generation WorksWhat RAG is and how it works: chunking, embeddings, vector search, hybrid search, reranking, GraphRAG and agentic RAG on AWS, Google and Azure.

Prompt engineering vs RAG vs fine-tuningWhen to use prompt engineering, retrieval-augmented generation, fine-tuning or continued pre-training, compared on cost, freshness, hallucination and privacy.