Microsoft AzureArtificial IntelligenceFOUNDATIONAL
Microsoft Azure AI Fundamentals Certification: Complete Guide 2026
AI-900
Microsoft Azure AI Fundamentals is Microsoft’s entry-level AI certification for anyone who wants to understand how AI solutions are built and delivered on Azure. Centered on exam AI-900, this credential is ideal for IT professionals, aspiring AI developers, data scientists, and ML engineers who want a clear Microsoft Azure AI Fundamentals overview before moving into more advanced Azure roles. The 60-minute exam includes 40-60 questions and requires a passing score of 700 out of 1000. You’ll be tested across five core areas: AI workloads and considerations (20%), ML principles on Azure (25%), computer vision (15%), NLP (25%), and generative AI workloads (15%). With no formal prerequisites and a strong focus on practical cloud AI concepts, Microsoft Azure AI Fundamentals is a smart starting point for building credibility in modern AI and Azure-focused careers.
A retail company wants to implement an AI solution to analyze customer purchase patterns and predict future sales. Which of the following workloads are most suitable for this scenario?
A
Machine learning
B
Computer vision
C
Natural language processing
D
Anomaly detection
E
Conversational AI
Show answer and explanation
Correct answers: A, D
Explanation
The most appropriate workloads for this scenario are machine learning and anomaly detection. Machine learning enables the creation of predictive models for sales forecasting based on historical data, while anomaly detection helps identify irregularities in purchase behavior that could affect the predictions. Other workloads, such as computer vision, natural language processing, and conversational AI, do not align with the company's objective of analyzing customer purchase patterns and predicting future sales.
A. Correct.
Machine learning is suitable because it enables the creation of predictive models to analyze customer purchase patterns and forecast future sales based on historical data.
B. Incorrect.
Computer vision is not suitable in this scenario because it focuses on analyzing visual data, such as images or videos, which is not relevant to the retail company's goal of analyzing purchase patterns.
C. Incorrect.
Natural language processing is not applicable here since this workload is primarily used to analyze and process text or speech data, not numerical purchase data.
D. Correct.
Anomaly detection is relevant because it can identify unusual trends or behaviors in customer purchases, which can complement the predictive analysis for sales forecasting.
E. Incorrect.
Conversational AI is not suitable for this scenario as it focuses on creating interactive systems like chatbots, which are unrelated to predicting sales or purchase patterns.
A company wants to automate the process of sorting incoming customer emails into categories such as 'Billing', 'Technical Support', and 'General Inquiry'. They also want to extract key details like customer names and account numbers from the emails. Which type of AI workload should the company consider?
A
Natural Language Processing (NLP)
B
Computer Vision
C
Anomaly Detection
D
Reinforcement Learning
Show answer and explanation
Correct answer: A
Explanation
The scenario described involves working with text data, specifically understanding and categorizing emails and extracting details like names and account numbers. These are tasks that fall under the domain of Natural Language Processing (NLP), which is designed to process and analyze human language in a way that computers can understand.
A. Correct.
Natural Language Processing (NLP) is the correct answer because it focuses on understanding and processing human language. Sorting emails into categories and extracting key details are common NLP tasks.
B. Incorrect.
Computer Vision is incorrect because it deals with analyzing and processing visual data, such as images and videos, rather than text.
C. Incorrect.
Anomaly Detection is incorrect because it focuses on identifying unusual patterns or behaviors in data, which is not relevant to categorizing emails or extracting text-based information.
D. Incorrect.
Reinforcement Learning is incorrect because it is a machine learning approach used for training agents to make decisions in dynamic environments through rewards and penalties, which is unrelated to the given scenario.
A company wants to implement a solution that uses Artificial Intelligence (AI) to analyze customer feedback and categorize it into positive, negative, or neutral sentiments. Which of the following AI workloads best aligns with this requirement?
A
Natural Language Processing (NLP)
B
Computer Vision
C
Anomaly Detection
D
Robotics Process Automation (RPA)
Show answer and explanation
Correct answer: A
Explanation
The task described requires analyzing customer feedback, which is a text-based dataset, to determine sentiment. This aligns with Natural Language Processing (NLP), a key AI workload that enables machines to understand and process human language.
A. Correct.
Natural Language Processing (NLP) is the correct choice because it focuses on analyzing and interpreting human language, such as text or speech. Sentiment analysis is a common NLP task.
B. Incorrect.
Computer Vision is incorrect because it deals with analyzing visual data like images or videos, which is not relevant to the task of analyzing customer feedback text.
C. Incorrect.
Anomaly Detection is incorrect because it is used to identify unusual patterns or outliers in data, such as detecting fraud, and does not align with the requirement of categorizing sentiment.
D. Incorrect.
Robotics Process Automation (RPA) is incorrect because it automates repetitive tasks and workflows, and it is not designed for analyzing or interpreting text sentiment.
Exam Content
Exam Domains & Topics
Master these 5 domains to pass your exam
1
Describe AI Workloads and Considerations
20%
2
Describe ML Principles on Azure
25%
3
Describe Computer Vision Workloads
15%
4
Describe NLP Workloads
25%
5
Describe Generative AI Workloads
15%
Who Should Take This Exam?
IT professionals seeking Microsoft Azure expertise
The Microsoft Azure AI Fundamentals (AI-900) certification validates foundational knowledge of artificial intelligence and machine learning concepts, and related Microsoft Azure services. This entry-level certification is ideal for individuals looking to demonstrate their understanding of common AI workloads and how to implement them on Azure, without requiring deep technical expertise in data science or software engineering.
Week 1
Foundation and AI Workload Concepts
Build foundational understanding of AI concepts, workloads, and responsible AI principles
Complete Microsoft Learn modules on AI fundamentals
Understand different types of AI workloads and their use cases
Master the six principles of responsible AI
Familiarize yourself with Azure portal and AI service locations
Week 2
Machine Learning Fundamentals
Deep dive into ML principles, Azure ML services, and model lifecycle
Understand supervised, unsupervised, and reinforcement learning
Learn about Azure Machine Learning workspace features
Practice with AutoML through Azure portal
Understand model evaluation metrics for different scenarios
Explore Azure ML Designer interface
Week 3
Computer Vision and NLP Workloads
Study Azure AI services for vision and language processing
Explore Computer Vision, Custom Vision, and Face services
Understand OCR and Form Recognizer capabilities
Study Language service features (sentiment, entity recognition, key phrases)
Learn about Speech service and Translator capabilities
Practice with conversational AI concepts
Week 4
Generative AI and Exam Preparation
Study generative AI concepts, review all domains, and take practice exams
Understand Azure OpenAI Service and its models
Learn about prompt engineering and content filtering
Review responsible AI considerations across all services
Complete multiple practice exams and identify weak areas
Review incorrect answers and fill knowledge gaps
Take official practice assessment
Study tips
Exam Format Understanding
Expect 40-60 questions including multiple choice, multiple select, drag-and-drop, and scenario-based questions
Questions are weighted differently - some count more than others, but you won't know which
You cannot return to previous questions once you move forward in certain sections
Flag questions you're unsure about for review if the section allows it
Read each question carefully - Microsoft exams often include subtle wording that changes the correct answer
Hands-On Practice
Create a free Azure account immediately and explore the services covered in the exam
Spend time in Azure Portal navigating to Computer Vision, Language, Speech, and ML services
Use the Microsoft Learn sandbox environments to complete hands-on labs without spending credits
Try the interactive demos on Azure AI services pages to understand capabilities without deployment
Practice with Azure Machine Learning Studio's AutoML and Designer features
Experiment with different Azure AI services using the free tier limits
Service Identification
Create a comparison chart of all Azure AI services with their specific use cases and capabilities
Memorize which service handles which task (e.g., Form Recognizer for document extraction, not Computer Vision)
Understand the difference between pre-built services and custom model services (Computer Vision vs Custom Vision)
Know when to use Azure OpenAI versus traditional Azure AI services
Be clear on Language service capabilities versus Speech service versus Translator
Study the decision tree: identify the AI workload type, then select the appropriate Azure service
Responsible AI Focus
Memorize the six principles: Fairness, Reliability & Safety, Privacy & Security, Inclusiveness, Transparency, Accountability
Understand how each principle applies to different AI scenarios and services
Be prepared for questions about ethical considerations in facial recognition, automated decision-making, and generative AI
Know Microsoft's approach to content filtering and safety in Azure OpenAI Service
Understand potential biases in ML models and how to address them
Review real-world examples where responsible AI principles are critical
Machine Learning Concepts
Clearly understand the difference between supervised (labeled data), unsupervised (unlabeled), and reinforcement learning
Know which algorithms solve which problems: regression (predict numbers), classification (predict categories), clustering (group similar items)
Memorize evaluation metrics: accuracy, precision, recall, F1-score for classification; RMSE, MAE, R-squared for regression
Understand the ML lifecycle: collect data → prepare/clean → train model → evaluate → deploy → monitor
Know what overfitting and underfitting mean and how to identify them
Understand the purpose of training, validation, and test datasets
Azure ML Studio Focus
Understand what AutoML does and when it's appropriate to use it versus custom solutions
Know the capabilities of Azure ML Designer (drag-and-drop visual interface)
Understand the concept of compute targets and when different types are needed
Know how to deploy models as web services for real-time or batch inference
Understand what an Azure ML workspace contains and its purpose
Be familiar with the model registry and versioning concepts
Effective Study Techniques
Follow Microsoft Learn paths in order as they build on each other progressively
Take notes while studying and create flashcards for services, definitions, and metrics
Watch video courses at 1.25-1.5x speed for efficiency, but pause to understand complex concepts
Schedule practice exams to simulate test conditions and identify weak areas
Review explanations for both correct and incorrect answers in practice tests
Join study groups or find an accountability partner preparing for the same exam
Teach concepts to others or explain them out loud to reinforce understanding
Final Week Preparation
Take at least 2-3 full-length practice exams under timed conditions
Review all flagged or incorrectly answered questions and understand why
Create a one-page summary sheet of all services, their purposes, and key features
Focus on your weakest domain based on practice exam results
Re-read the official exam skills outline and ensure you understand every bullet point
Don't cram new material the day before - focus on reviewing and reinforcing existing knowledge
Get adequate sleep before exam day - mental clarity is crucial for scenario-based questions
Exam day checklist
Arrive 15 minutes early if taking the exam at a test center, or log in 15 minutes early for online proctoring
Bring valid, government-issued ID that exactly matches your exam registration name
Read the exam instructions and NDA carefully before starting the timer
Read each question completely before looking at the answer choices to avoid bias
Eliminate obviously wrong answers first, then choose between remaining options
For scenario-based questions, identify the key requirements and constraints before selecting an answer
Pay attention to qualifiers like 'most', 'least', 'best', 'primary', as they change which answer is correct
Don't spend more than 1-2 minutes on any single question initially - flag difficult ones and move on
If a question asks about 'responsible AI', focus on the six principles in your answer selection
For service identification questions, think about the specific capability needed, not just the general category
Manage your time: with 60 minutes and up to 60 questions, you have about 1 minute per question
Review your flagged questions if time permits, but trust your first instinct unless you find clear evidence to change
Stay calm and confident - this is a foundational exam testing broad knowledge, not deep technical expertise
Career
Career Opportunities
Roles and salary potential for Microsoft Azure AI Fundamentals certified professionals
Related Job Titles
AI DeveloperData ScientistML Engineer
$90,000
Average Annual Salary
From the Blog
Related Articles
Guides and insights for Microsoft Azure AI Fundamentals professionals
There are no strict formal prerequisites for the Microsoft Azure AI Fundamentals certification. However, Microsoft Azure recommends having foundational knowledge of artificial intelligence concepts and some hands-on experience before attempting the exam. Candidates who invest time in study materials and practice exams typically perform best.
FAQ
Microsoft Azure AI Fundamentals FAQs
Common questions about the AI-900 certification exam
The Microsoft Azure AI Fundamentals is a professional certification offered by Microsoft Azure that validates your expertise in the relevant technology domain. The exam code is AI-900. This certification demonstrates your ability to design, implement, and manage solutions using Microsoft Azure technologies.
The Microsoft Azure AI Fundamentals exam typically contains 40-60 questions. These questions are a mix of multiple-choice and scenario-based questions designed to test both theoretical knowledge and practical application.
The passing score for the Microsoft Azure AI Fundamentals exam is 700/1000. Note that Microsoft Azure uses a scaled scoring system, so focus on understanding all exam domains thoroughly rather than just achieving the minimum score.
The Microsoft Azure AI Fundamentals exam duration is 60 minutes (1 hours). This includes time for reviewing your answers. We recommend practicing with timed mock exams to manage your time effectively.
The Microsoft Azure AI Fundamentals exam costs $99. Prices may vary by region and are subject to change. Microsoft Azure occasionally offers discounts or voucher programs for certification exams.
The Microsoft Azure AI Fundamentals certification is valid for Lifetime. To maintain your certification, you'll need to recertify before it expires, either by passing the current exam version or through Microsoft Azure's continuing education program.
While Microsoft Azure doesn't always require formal prerequisites, we recommend having hands-on experience with the relevant technologies. Familiarity with core concepts and practical experience will significantly improve your chances of passing the exam.
Yes, the Microsoft Azure AI Fundamentals exam is proctored and can be taken either at a testing center or online through remote proctoring. Online proctoring allows you to take the exam from home while being monitored via webcam. Ensure you have a quiet, private space with a stable internet connection if choosing the online option.
If you don't pass the Microsoft Azure AI Fundamentals exam on your first attempt, you can retake it. Microsoft Azure typically has a waiting period between attempts (usually 14 days for the first retake). Use this time to review the areas where you struggled and take additional practice exams.
To prepare for the Microsoft Azure AI Fundamentals exam, we recommend: 1) Review the official exam guide and objectives, 2) Gain hands-on experience with the technologies, 3) Use practice exams to identify knowledge gaps, 4) Study each exam domain thoroughly, and 5) Join study groups or forums to discuss challenging topics with other candidates.
About the Microsoft Azure AI Fundamentals Certification
The Microsoft Azure AI Fundamentals (AI-900) is a foundational-level certification offered by Microsoft Azure. This certification validates your expertise in artificial intelligence and is recognized globally by employers seeking qualified professionals. The exam consists of 40-60 questions to be completed in 60 minutes, with a passing score of 700/1000. The exam fee is $99, and the certification is valid for Lifetime.
Why Get Microsoft Azure AI Fundamentals Certified?
Career Advancement: Certified professionals earn an average of $90,000 per year. Microsoft Azure-certified professionals are among the most sought-after in the artificial intelligence industry.
Industry Recognition: Microsoft Azure certifications are respected worldwide by employers, demonstrating verified competency in artificial intelligence technologies and practices.
Skill Validation: The Microsoft Azure AI Fundamentals exam rigorously tests your knowledge across 5 domains, ensuring you have the practical skills employers demand.
Microsoft Azure AI Fundamentals Exam Format & Details
The AI-900 exam is designed to test both theoretical knowledge and practical application. Candidates are given 60 minutes to complete the exam, which contains approximately 40-60 questions. A score of 700/1000 is required to pass. As a foundational-level exam, it focuses on core concepts and terminology, making it accessible to professionals new to the field.
Exam Domains & Topics
The Microsoft Azure AI Fundamentals exam covers 5 key domains. Understanding the weight of each domain helps you allocate your study time effectively:
Describe AI Workloads and Considerations (20% of exam)
Describe ML Principles on Azure (25% of exam)
Describe Computer Vision Workloads (15% of exam)
Describe NLP Workloads (25% of exam)
Describe Generative AI Workloads (15% of exam)
Who Should Take the Microsoft Azure AI Fundamentals Exam?
This certification is designed for professionals in the following roles:
IT professionals seeking Microsoft Azure expertise
Artificial Intelligence practitioners looking to validate their skills
Professionals preparing for a career in artificial intelligence
Technical specialists aiming to advance their career with an industry-recognized credential
Team leads and managers who need to understand artificial intelligence concepts
Career Opportunities & Salary
Earning the Microsoft Azure AI Fundamentals certification opens doors to roles such as AI Developer, Data Scientist, ML Engineer. Certified professionals earn an average salary of $90,000 per year, reflecting the high demand for artificial intelligence skills in today's job market.
Recertification & Renewal
The Microsoft Azure AI Fundamentals certification is valid for Lifetime. To maintain your credential, you will need to meet Microsoft Azure's renewal requirements before your certification expires. This may include earning continuing education credits, passing a recertification exam, or earning a higher-level certification.
Exam Registration & Cost
The AI-900 exam costs $99. You can register through Microsoft Azure's official website or an authorized testing center. Most candidates choose between in-person testing at a Pearson VUE or PSI center and online proctored exams taken from home. Be sure to review the exam policies, including identification requirements and prohibited items, before your test date.
How to Prepare for AI-900
Most candidates need 4-8 weeks of dedicated study to prepare for the Microsoft Azure AI Fundamentals exam. Start by reviewing the official exam objectives, then work through each domain systematically. Regular practice with exam-style questions is essential for building confidence and identifying weak areas. Combine reading with hands-on practice to develop both theoretical knowledge and practical skills.
HydraNode publishes 285 free AI-900 practice questions with answers and explanations, plus a timed practice exam drawn from the same bank. Every question is written to the published objectives, so what you practise matches the format and difficulty of the actual AI-900 exam.
