
There are now more AI certifications than ever, and the choice can feel overwhelming. The good news: you only need to answer two questions to pick the right first one.
Question 1: Which cloud does your target employer use?
Look at ten job posts for the role you want. Count how often AWS, Azure and Google Cloud appear. Pick the cloud that shows up most. If it is a tie, AWS and Azure have the largest number of openings in most markets.
Question 2: Do you write code?
If not, start with a fundamentals or leader level exam. If you already build applications, go straight to an associate exam. Employers value associate level certifications much more.

Our recommendations
| Your situation | Start with | Then |
|---|---|---|
| Non technical or business role, any cloud | Google Cloud Generative AI Leader or AWS AI Practitioner (AIF-C01) | A cloud fundamentals exam |
| New to cloud and AI, Microsoft shop | AI-901 Azure AI Fundamentals | AI-103 |
| Developer, Microsoft shop | AI-103 Azure AI Apps and Agents Developer | AI-300 for MLOps |
| New to cloud and AI, AWS shop | AWS AI Practitioner (AIF-C01) | Machine Learning Engineer Associate |
| Experienced developer, AWS shop | AWS Generative AI Developer Professional (AIP-C01) | Solutions Architect Associate if you lack cloud basics |
| Platform or DevOps engineer | Certified Kubernetes Administrator (CKA) | A cloud associate exam |

Three mistakes to avoid
- Studying for a retired exam. AI-102, AI-900 and DP-100 retired in 2026. Check the exam page before you buy any material.
- Collecting fundamentals badges. One fundamentals exam is useful. Three is not. Move to associate level as soon as you can.
- Skipping hands on practice. Modern AI exams are scenario based. If you have never built a RAG app or an agent, the questions are much harder.
Try a free practice quiz first
Every exam below has free practice questions on its page, each with a full explanation, so you can see which one suits you before committing.