Microsoft Certified: Azure AI Fundamentals Practice Exam 2025: Latest Questions

You are designing an AI solution that must explain to users why a loan application was approved or denied. Which responsible AI principle is MOST directly addressed by providing clear, human-understandable reasons for the model’s decision?

A team wants to predict tomorrow’s energy consumption (a numeric value) using historical sensor readings. Which type of machine learning problem is this?

You need to extract printed text from scanned invoices and store the text for searching. Which Azure AI service capability is the BEST fit?

A support chatbot must identify whether a customer message expresses a positive, negative, or neutral opinion. Which NLP workload is this?

You are training a model in Azure Machine Learning and notice it performs extremely well on training data but poorly on new, unseen data. What issue is MOST likely occurring?

A company plans to use an AI model to recommend medical follow-up actions. They must ensure the system behaves consistently under expected conditions and has a clear fallback when confidence is low. Which responsible AI principle BEST aligns with this requirement?

You need to build a solution that can locate and label multiple products on a store shelf photo by drawing bounding boxes around each product. Which computer vision capability should you use?

You want to analyze customer feedback emails to automatically extract key topics and the most important phrases mentioned. Which NLP feature is MOST appropriate?

You trained a classification model and it achieves 98% accuracy. However, only 2% of transactions are actually fraudulent, and the model misses most fraud cases. Which evaluation approach would BEST help you understand this problem?

A global organization wants to make recorded training videos searchable. Users should be able to search for a phrase and jump to the moment it was spoken. Which Azure AI capability is REQUIRED to enable this scenario?

You want to build a model that predicts whether a customer will cancel a subscription (Yes/No) based on historical customer data. What type of machine learning task is this?

You are analyzing customer support tickets and want to automatically identify the language of each ticket before routing it to an agent. Which Azure AI capability best fits this requirement?

A retailer wants to reduce the risk of biased outcomes when using AI to recommend products. Which action is a best practice for responsible AI?

You are using Azure Machine Learning to train a model. You notice the model performs very well on the training dataset but poorly on new data. What is the most likely issue?

You plan to use Azure AI Vision to identify common objects in photos (for example, "car", "tree", "person"). Which feature should you use?

A support team wants a chatbot that can answer common questions using a curated knowledge base of internal policy documents. Which approach is most appropriate?

You have a dataset with multiple numeric features and no label column. You want to group similar records to identify customer segments. What type of machine learning should you use?

You are building an app that extracts key-value pairs (for example, InvoiceNumber, TotalDue) from many invoice formats, including scanned images and PDFs. Which Azure service is designed for this scenario?

A company deploys a sentiment analysis model for social media posts. Over time, the model accuracy decreases because the language and slang used by customers changes. What should the company do to address this?

You are designing a solution to translate live spoken conversations between English and Japanese in near real time. The solution must convert speech to text, translate it, and optionally speak the translation aloud. Which Azure service best aligns to this end-to-end requirement?

A retail website wants to show product images in search results with tags such as "shoe", "backpack", and "outdoor" without training a custom model. Which Azure service should they use?

You are building a chatbot to answer questions from a large set of internal policy documents. Users should receive a concise answer with citations to the source passages. Which approach best meets this requirement?

A data scientist creates a model with very high accuracy on the training dataset, but its performance drops significantly on new data. Which issue is most likely occurring?

A manufacturer wants to detect small defects on products using images. Only 120 labeled defect images are available, and collecting more will take weeks. What is the best initial approach to improve model performance while minimizing new labeling effort?

A healthcare organization is concerned that an AI system might produce different outcomes for patients based on sensitive attributes such as race or gender. Which responsible AI principle is MOST directly related to this concern?