Practice for the AI-900: Microsoft Azure AI Fundamentals exam with 13 exam-style practice questions, instant answer reveals, and concise explanations of every correct answer. Topics include: A banking app uses facial recognition to verify users during transactions. The system compares a live selfie to a govern. Follow @CertPunch and visit certpunch.com for more certification practice exams and study content.
Prefer hands-on? Take this round as an interactive practice test — answer every question, get instant feedback, and see your score: Start the AI-900: Microsoft Azure AI Fundamentals practice test →
What you will practice
- A banking app uses facial recognition to verify users during transactions. The system compares a live selfie…
- When engineering prompts for a large language model, a developer wants to ensure the output is both high-qual…
- A computer vision system must process a colorful photograph before any feature extraction takes place. How is…
- A global corporation needs to implement real-time translation during international video conferences while ma…
- A research organization needs to implement semantic search across multiple languages for their scientific doc…
- Modern multi-modal models are used to generate captions for images by integrating visual and textual data. Wh…
Answers and explanations
Tap a question to expand the answer and the exam reasoning. Try to commit to your own pick first.
Q1. A banking app uses facial recognition to verify users during transactions. The system compares a live selfie to a government ID photo. Which Azure Face API operation is being used?
Answer: D. Face Verification
Face Verification performs a one-to-one comparison to check if two faces belong to the same person. Use this as your exam cue: authentication scenarios comparing a live selfie to an ID require verification, not identification.
Q2. When engineering prompts for a large language model, a developer wants to ensure the output is both high-quality and aligned with specific requirements. Which approach best enhances the model's performance?
Answer: C. Supplying detailed instructions and a clear output indicator while including relevant contextual information.
Providing detailed instructions, clear output indicators, and relevant context explicitly guides the model to generate high-quality responses. Expecting the model to infer tasks from raw data often leads to unpredictable and poorly aligned results.
Q3. A computer vision system must process a colorful photograph before any feature extraction takes place. How is a color image typically represented in numerical form for computer vision algorithms?
Answer: C. As three separate two-dimensional matrices—one each for red, green, and blue—with values typically ranging from 0 to 255.
Computer vision algorithms typically represent color images as three two-dimensional matrices corresponding to red, green, and blue channels. Values range from zero to two hundred fifty-five, whereas a single matrix represents a grayscale image.
Q4. A global corporation needs to implement real-time translation during international video conferences while maintaining high security standards. Which combination of Azure AI services and deployment options would BEST meet these requirement…
Answer: C. Azure AI Speech with edge containers and speech translation
Azure AI Speech provides the speech translation capability required for real-time spoken language translation. Deploying via edge containers meets high security standards by processing sensitive audio data locally rather than in the cloud.
Q5. A research organization needs to implement semantic search across multiple languages for their scientific documents. Which sequence of Knowledge Mining capabilities would BEST address this requirement?
Answer: B. OCR → Translation → Entity Recognition → Indexing
Knowledge mining pipelines first use cognitive skills to enrich data before indexing. OCR extracts text from images, translation standardizes the language, and entity recognition extracts concepts before everything is finally indexed for search.
Q6. Modern multi-modal models are used to generate captions for images by integrating visual and textual data. Which approach best explains how these models connect images and text?
Answer: D. By learning a shared representation where an image encoder extracts visual features and a language encoder generates text embeddings, mapping both into a common vector space.
Multi-modal models connect images and text by mapping outputs from visual and language encoders into a shared vector space. Relying only on metadata or optical character recognition fails to capture the full semantic meaning of the visual scene.
Q7. A global manufacturing company needs to translate technical documentation while maintaining specific industry terminology and formatting. Which Azure AI Translator capability and method would be MOST appropriate?
Answer: C. Asynchronous batch translation with custom models
Asynchronous batch document translation handles large files while preserving original formatting. Pairing this with custom translation models ensures that specific industry terminology and domain-specific vocabulary are translated accurately.
Q8. An enterprise is looking to deploy a customized AI solution within Azure. Which feature of Azure AI Foundry is most critical for tailoring a pre-trained model to meet a specific business context?
Answer: B. The ability to fine tune models and incorporate a proprietary knowledge base.
The ability to fine-tune models allows an enterprise to adapt pre-trained base models using proprietary domain data. Incorporating a proprietary knowledge base further grounds the model to ensure responses fit the specific business context.
Q9. A language learning platform needs to provide feedback on students' pronunciation in multiple languages. Which combination of Azure AI services and features would BEST meet this requirement?
Answer: B. Azure AI Speech with pronunciation assessment and neural voices
Azure AI Speech includes pronunciation assessment to analyze spoken language and provide feedback. For the exam, remember that translation services map languages, while the speech service handles audio, including pronunciation and fluency grading.
Q10. A retail company wants to implement real-time fraud detection in their e-commerce platform. They have limited AI expertise but need a solution that can handle complex transaction patterns. Which Azure AI approach would be MOST suitable?
Answer: C. Use prebuilt Anomaly Detector service from Azure AI Metrics Advisor
Using a prebuilt anomaly detection service is ideal for real-time pattern analysis without requiring deep machine learning expertise. Building custom models in Azure Machine Learning would require significant AI knowledge, directly contradicting the limited expertise constraint.
Q11. A company processes insurance claims using scanned forms with unique layouts specific to their workflow. They need to extract "Policy Number" and "Claim Date" fields reliably. Which Azure service should they use?
Answer: A. Azure Document Intelligence (custom model)
A custom model allows the organization to train the system on their specific forms, capturing unique layouts and reliably extracting critical fields such as policy number and claim date. Prebuilt models are better suited for standardized documents like invoices or passports.
Q12. A travel booking chatbot frequently misinterprets customer requests. The development team wants to improve its understanding of user intentions. Which sequence of steps would BEST address this issue?
Answer: C. Author intents → Label utterances → Train model → Test accuracy
The correct sequence begins by defining what the chatbot should understand through authoring intents, followed by labeling sample utterances to create a robust training dataset. You must always train the model after authoring intents and labeling utterances, not before.
Q13. When working with Azure AI services for natural language processing (NLP), which method allows both developers and non-developers to interact with and test NLP models without writing code?
Answer: D. Utilizing studio interfaces such as the Language Studio
Studio interfaces like Language Studio provide a graphical environment that enables both developers and non-developers to test natural language processing models without writing code. For the exam, remember that application programming interfaces and client libraries require programming knowledge.
More AI-900: Microsoft Azure AI Fundamentals drills and other practice exams are on @CertPunch. New rounds drop every few days at certpunch.com.