DP-100 Is Retired: What Replaced Azure’s Data Science Certification

If you were studying for DP-100 (Azure Data Scientist Associate), the short version: the exam retired on June 1, 2026, as part of Microsoft’s sweeping 2026 reshuffle of its AI certification line. Its successor is AI-300, which leads to a new credential — Microsoft Certified: Machine Learning Operations Engineer Associate. Here’s what changed, what it means for a cert you already hold, and what to do if you were mid-study.

What AI-300 actually covers

AI-300 is not DP-100 with a new number — the role itself was redefined. Where DP-100 centered on the data scientist building and training models, AI-300 targets what Microsoft calls AI operations (AIOps): running machine learning and generative AI in production. Per Microsoft’s official exam page, the five assessed domains are:

• Design and implement an MLOps infrastructure
• Implement machine learning model lifecycle and operations
• Design and implement a GenAIOps infrastructure
• Implement generative AI quality assurance and observability
• Optimize generative AI systems and model performance

The toolchain reflects that: Azure Machine Learning and Microsoft Foundry, plus DevOps machinery — GitHub Actions, infrastructure-as-code with Bicep, and the Azure CLI. Microsoft pitches it at intermediate level, expects a data-science background with Python, and gives you 120 minutes in a proctored exam. Notably, as of this writing Microsoft’s own practice assessment for AI-300 isn’t available yet — official prep material is still catching up to the exam.

I hold DP-100 — is it worthless now?

No. A retirement stops new exams; it doesn’t erase earned credentials. Your certification remains on your Microsoft transcript, and employers understand that a retired exam was legitimate when you passed it. The practical wrinkle is renewal — retired certifications eventually age out of the renewal cycle, so check your specific renewal window in your Microsoft Learn certification dashboard rather than assuming either way.

If you were mid-study for DP-100

Your work isn’t wasted — it forks. Decide by role, not sunk cost:

• You run models in production (or want to): AI-300 is your path — your Azure ML knowledge carries straight over, and you’ll add the GenAIOps and DevOps layers.
• You build AI applications and agents: look at AI-103 instead — the new Azure AI Apps and Agents Developer credential is the builder’s fork of this reshuffle.
• You’re earlier in the journey: AI-901 fundamentals is the modern on-ramp before any associate exam.

The 2026 pattern, in one paragraph

DP-100’s retirement wasn’t an isolated event: AI-900 retired June 30 (replaced by AI-901), DP-100 retired June 1 (replaced by AI-300), the brand-new AI-103 launched for generative-AI and agent developers — and AI-102, contrary to many roundups, was not retired. The full map is in Microsoft AI certifications in 2026, and the AI-102/AI-103 decision gets its own breakdown in AI-102 vs AI-103.

We don’t have an AI-300 practice bank yet — it’s on our evaluation list as the exam matures. Meanwhile, all eleven of our live certification banks, including AI-901 and AI-103, are at the AI exam prep hub.