MICROSOFT CERTIFICATION
AI-900 Azure AI Fundamentals Practice Exam
Exam Number: 3102 | Last updated 16-Apr-26 | 800+ questions across 4 vendor-aligned objectives
The AI-900 Azure AI Fundamentals certification validates the skills of professionals seeking foundational knowledge of artificial intelligence and machine learning concepts on Azure. This exam measures your ability to work with Azure AI services, Azure Machine Learning, Azure AI Document Intelligence, Azure AI Vision, Azure AI Language, demonstrating both conceptual understanding and practical implementation skills required in today’s enterprise environments.
The heaviest exam domains include Describe Fundamental Principles of Machine Learning on Azure (20–25%), Describe Artificial Intelligence Workloads and Considerations (15–20%), and Describe Features of Computer Vision Workloads on Azure (15–20%). These areas collectively represent the majority of exam content and require focused preparation across their respective subtopics.
Additional domains tested include Describe Features of Natural Language Processing Workloads on Azure (15–20%), and Describe Features of Generative AI Workloads on Azure (15–20%). Together, these areas round out the full exam blueprint and ensure candidates possess well-rounded expertise across the certification scope.
Every answer links to the source. Each explanation below includes a hyperlink to the exact Microsoft documentation page the question was derived from. PowerKram is the only practice platform with source-verified explanations. Learn about our methodology →
828
practice exam users
94.4%
satisfied users
88.4%
passed the exam
4.8/5
quality rating
Test your AI-900 Azure AI Fundamentals knowledge
10 of 800+ questions
Question #1 - Describe Fundamental Principles of Machine Learning on Azure
A marketing team at a retailer wants to predict customer churn using historical purchase data but has no coding expertise.
Which Azure ML capability enables model building without code?
A) Azure Databricks with PySpark
B) Azure HDInsight Spark clusters
C) Azure Synapse SQL pools
D) Azure ML Designer with drag-and-drop
Show solution
Correct answers: D – Explanation:
Azure ML Designer provides a visual drag-and-drop interface requiring no coding. Databricks and PySpark need programming. Synapse SQL is for data warehousing. HDInsight requires cluster management and coding. Source: Check Source
Question #2 - Describe Fundamental Principles of Machine Learning on Azure
A marketing team at a retailer wants to predict customer churn using historical purchase data but has no coding expertise.
Which Azure ML capability enables model building without code?
A) Azure HDInsight Apache Spark cluster with Scala-based training scripts
B) Azure Machine Learning Designer with a visual drag-and-drop pipeline interface
C) Azure Synapse Analytics dedicated SQL pool with stored procedure models
D) Azure Databricks workspace with PySpark notebook-based model development
Show solution
Correct answers: B – Explanation:
Azure ML Designer provides a visual drag-and-drop interface requiring no coding. Databricks and PySpark require Python or Scala programming skills. Synapse SQL pools handle data warehousing queries, not ML model building. HDInsight with Spark requires cluster management and coding expertise. Source: Check Source
Question #3 - Describe Fundamental Principles of Machine Learning on Azure
HR wants to predict employee salaries based on experience, education, and role. The target is a continuous number.
What type of machine learning task is this?
A) Classification that predicts which discrete salary band the employee falls into
B) Regression that predicts a continuous numeric salary value from the input features
C) Anomaly detection that identifies employees with salaries deviating from norms
D) Clustering that groups employees into natural salary segments without labels
Show solution
Correct answers: B – Explanation:
Regression predicts continuous numeric values like salary amounts from input features. Classification predicts discrete categories or labels, not continuous values. Clustering groups similar data points without a predefined target variable. Anomaly detection identifies unusual data points that deviate from established patterns. Source: Check Source
Question #4 - Describe Artificial Intelligence Workloads and Considerations
A bank’s AI loan system disproportionately denies loans to certain demographics despite comparable financial profiles.
Which responsible AI principle is primarily violated?
A) Fairness requiring the model to produce equitable outcomes across demographic groups
B) Privacy and security requiring the model to protect applicant personal data properly
C) Transparency requiring the model to explain each individual decision to applicants
D) Reliability and safety requiring the model to operate consistently under all conditions
Show solution
Correct answers: A – Explanation:
Fairness requires equitable treatment across groups. The model exhibits bias by treating comparable financial profiles differently based on demographic characteristics. Transparency relates to whether decisions can be explained, not whether they are equitable. Privacy concerns data protection practices. Reliability refers to consistent, predictable operation under various conditions. Source: Check Source
Question #5 - Describe Artificial Intelligence Workloads and Considerations
A city evaluates AI workloads for public services and needs to identify which scenario is a computer vision workload.
Which is an example of a computer vision AI workload?
A) Generating automated response drafts for citizen email inquiries to city offices
B) Analyzing traffic camera footage to automatically count vehicles at intersections
C) Predicting next month water consumption volume from historical municipal utility data
D) Translating public announcements from English into twelve additional regional languages
Show solution
Correct answers: B – Explanation:
Analyzing camera footage to count vehicles extracts structured information from visual data — a core computer vision task. Language translation is a natural language processing workload. Predicting consumption from historical data is a standard machine learning forecasting task. Generating email responses is a natural language generation workload. Source: Check Source
Question #6 - Describe Features of Computer Vision Workloads on Azure
A real estate company wants to auto-tag property photos with labels like “kitchen” and “swimming pool” for search.
Which Azure AI Vision capability should they use?
A) Optical Character Recognition to extract any visible text and signage from listing photos
B) Face detection and analysis to identify persons appearing in property listing images
C) Image tagging through the Image Analysis API to generate descriptive content labels
D) Custom Vision object detection to draw bounding boxes around features in each photo
Show solution
Correct answers: C – Explanation:
Image tagging through the Image Analysis API automatically generates descriptive tags for recognized objects and scenes — ideal for categorizing property photos at scale. OCR extracts visible text characters, not visual scene descriptions. Face detection identifies human faces, not property features like pools or kitchens. Custom Vision requires training with labeled data, which is unnecessary when pre-built tags cover the needed categories. Source: Check Source
Question #7 - Describe Features of Computer Vision Workloads on Azure
A library digitizes historical documents by scanning. Scanned images need text extracted for online search.
Which service extracts text from scanned document images?
A) Azure AI Vision Read API performing optical character recognition on the images
B) Azure AI Language key phrase extraction applied to the recognized document content
C) Azure AI Content Safety service scanning the document images for policy violations
D) Azure AI Translator document translation service converting scanned page content
Show solution
Correct answers: A – Explanation:
The Azure AI Vision Read API performs optical character recognition on images to extract printed and handwritten text, making scanned documents searchable. Key phrase extraction operates on existing text strings, not raw images. Content Safety evaluates images for harmful content, not text extraction. Translator converts text between languages but requires text input, not scanned images. Source: Check Source
Question #8 - Describe Features of Natural Language Processing Workloads on Azure
A hotel chain wants to analyze guest reviews to understand satisfaction by specific aspects like room quality and cleanliness.
Which feature provides aspect-level sentiment detail?
A) Named Entity Recognition that identifies and categorizes mentioned entities like locations
B) Language detection that identifies which language each review was originally written in
C) Text summarization that condenses long reviews into shorter representative passages
D) Sentiment analysis with opinion mining that evaluates sentiment toward specific aspects
Show solution
Correct answers: D – Explanation:
Opinion mining goes beyond overall sentiment to identify specific aspects (targets) mentioned in text and the sentiment expressed about each one. Named Entity Recognition identifies entities like names and places but does not assess emotional tone. Language detection identifies which language text is written in. Text summarization condenses content but does not evaluate sentiment toward specific aspects. Source: Check Source
Question #9 - Describe Features of Natural Language Processing Workloads on Azure
An international non-profit receives requests in dozens of languages and must identify each language before routing to specialized teams.
Which capability should be the first processing step?
A) Custom text classification that categorizes each request by department and topic
B) Text translation that converts all incoming requests into a single standard language
C) Language detection that identifies which language each individual request is written in
D) Key phrase extraction that identifies the most important terms within each request
Show solution
Correct answers: C – Explanation:
Language detection identifies the language of input text, which is the necessary first step before routing to language-specific teams for processing. Translation converts text but the organization wants to route to native-speaking teams, not translate everything. Key phrase extraction identifies important terms but does not determine the language. Custom classification categorizes content by topic, not by language. Source: Check Source
Question #10 - Describe Features of Generative AI Workloads on Azure
A software company integrates Azure OpenAI to suggest code completions from comments and partial code.
Which generative AI characteristic enables this?
A) The ability to generate new content based on patterns learned from large training datasets
B) The ability to detect anomalies and errors within existing structured code repositories
C) The ability to classify input text into a predefined set of programming language categories
D) The ability to recognize visual elements and diagrams in code architecture screenshots
Show solution
Correct answers: A – Explanation:
Generative AI models produce new content — including code — based on patterns learned during training, enabling contextually appropriate code suggestions. Classification assigns labels to input but does not generate new output content. Anomaly detection identifies outliers in existing data. Image recognition processes visual input, not code generation tasks. Source: Check Source
Get 800+ more questions with source-linked explanations
Every answer traces to the exact Microsoft documentation page — so you learn from the source, not just memorize answers.
Exam mode & learn mode · Score by objective · Updated 16-Apr-26
Learn more...
What the AI-900 Azure AI Fundamentals exam measures
- Describe Artificial Intelligence Workloads and Considerations (15–20%) — Evaluate your ability to implement and manage tasks within this domain, including real-world job skills and scenario-based problem solving.
- Describe Fundamental Principles of Machine Learning on Azure (20–25%) — Evaluate your ability to implement and manage tasks within this domain, including real-world job skills and scenario-based problem solving.
- Describe Features of Computer Vision Workloads on Azure (15–20%) — Evaluate your ability to implement and manage tasks within this domain, including real-world job skills and scenario-based problem solving.
- Describe Features of Natural Language Processing Workloads on Azure (15–20%) — Evaluate your ability to implement and manage tasks within this domain, including real-world job skills and scenario-based problem solving.
- Describe Features of Generative AI Workloads on Azure (15–20%) — Evaluate your ability to implement and manage tasks within this domain, including real-world job skills and scenario-based problem solving.
How to prepare for this exam
- Review the official exam guide to understand every objective and domain weight before you begin studying
- Complete the relevant Microsoft Learn learning path to build a structured foundation across all exam topics
- Get hands-on practice in an Azure free-tier sandbox or trial environment to reinforce what you have studied with real configurations
- Apply your knowledge through real-world project experience — whether at work, in volunteer roles, or contributing to open-source initiatives
- Master one objective at a time, starting with the highest-weighted domain to maximize your score potential early
- Use PowerKram learn mode to study by individual objective and review detailed explanations for every question
- Switch to PowerKram exam mode to simulate the real test experience with randomized questions and timed conditions
Career paths and salary outlook
Earning this certification can open doors to several in-demand roles:
- AI Solutions Consultant: $95,000–$130,000 per year (based on Glassdoor and ZipRecruiter data)
- Cloud Solutions Associate: $80,000–$110,000 per year (based on Glassdoor and ZipRecruiter data)
- Technical Sales Specialist – AI: $90,000–$125,000 per year (based on Glassdoor and ZipRecruiter data)
Official resources
Microsoft provides comprehensive free training to prepare for the AI-900 Azure AI Fundamentals exam. Start with the official Microsoft Learn learning path for structured, self-paced modules covering every exam domain. Review the exam study guide for the complete skills outline and recent updates.