This is the free Microsoft AB-731 practice question bank —
50 of 99 total questions, each with a full explanation, free to
read with no signup required. Updated 2026-07-09.
Every answer is verified against official Microsoft documentation —
see our methodology.
Question 1
What is a key feature of Microsoft 365 Copilot that aligns with the Microsoft responsible AI principles of transparency, reliability, and safety?
A. Automatically approves AI-generated content for company-wide publishing.
B. Removes the need for a human to review AI-generated outputs.
C. Provides grounded, verifiable responses based on organizational data.
D. Enables users to select from an authorized catalog of AI models.
Show Answer
Correct Answer: C
Explanation: Microsoft 365 Copilot grounds responses in organizational data and provides citations/references where applicable, supporting transparency, reliability, and safety by enabling users to verify AI-generated content. The other options incorrectly imply autonomous approval, removal of human oversight, or describe a different capability rather than a core responsible AI feature.
Question 2
Your company uses Microsoft 365 Copilot.
You identify several business processes that require custom workflows and specialized automation.
You need to recommend a solution that extends Copilot capabilities while minimizing development effort and costs.
What should you recommend?
A. Create an agent by using Microsoft Foundry.
B. Deploy a Copilot connector.
C. Build a custom agent by using the full experience of Microsoft Copilot Studio.
D. Build a declarative agent by using the lite experience of Microsoft Copilot Studio.
Show Answer
Correct Answer: D
Explanation: Declarative agents built with the lite experience of Microsoft Copilot Studio are intended to extend Microsoft 365 Copilot with custom business workflows while requiring minimal development effort and lower cost. The full experience is better suited for more complex, highly customized agents, Copilot connectors expose external data rather than implementing workflow automation, and Microsoft Foundry is not the recommended low-effort option for extending Microsoft 365 Copilot in this scenario.
Question 3
DRAG DROP
-
Match the Microsoft responsible AI principles to the appropriate descriptions.
To answer, drag the principle from the column on the left to its description on the right. Each principle may be used once, more than once, or not at all.
NOTE: Each correct match is worth one point.
Show Answer
Correct Answer: Privacy and security
Transparency
Inclusiveness
Accountability
Explanation: These map directly to Microsoft's Responsible AI principles: protecting data is Privacy and security; explaining decisions is Transparency; accessibility is Inclusiveness; and human oversight is Accountability.
Question 4
HOTSPOT
-
For each of the following statements, select Yes if the statement is true. Otherwise, select No.
NOTE: Each correct selection is worth one point.
Show Answer
Correct Answer: Yes
No
Yes
Explanation: Copilot drafts professional emails and summarizes conversations. It assists with Excel but does not autonomously update spreadsheets without user action. It can summarize recent updates across Microsoft 365 documents, spreadsheets, and presentations using Microsoft Graph context.
Question 6
HOTSPOT
-
For each of the following statements, select Yes if the statement is true. Otherwise, select No.
NOTE: Each correct selection is worth one point.
Show Answer
Correct Answer: Yes
No
No
Explanation: Azure AI Language provides sentiment analysis. Text translation is handled by Azure AI Translator, and OCR/text extraction from scanned documents is provided by Azure AI Vision or Azure AI Document Intelligence, not Azure AI Language.
Question 7
HOTSPOT
-
Select the answer that correctly completes the sentence.
Show Answer
Correct Answer: data security
Explanation: The principle of least privilege enforces access control so Copilot can only access data the user is authorized to view, which is a data security control.
Question 8
Your company plans to use generative AI to help summarize and analyze internal business documents.
You need to recommend a solution to prevent generative AI from accessing confidential or classified information.
What should you include in the recommendation?
A. an information barrier (IB) policy
B. communication monitoring
C. a data retention policy
D. data governance
Show Answer
Correct Answer: D
Explanation: Data governance is the broad set of policies and controls used to classify, label, protect, and control access to organizational data. For generative AI, strong data governance ensures confidential or classified information is appropriately restricted and not available for AI grounding or retrieval by unauthorized users. The other options address specific compliance or communication scenarios but do not comprehensively prevent AI from accessing sensitive information.
Question 9
HOTSPOT
-
Select the answer that correctly completes the sentence.
Show Answer
Correct Answer: tokens
Explanation: LLM usage is typically billed based on the number of input and output tokens processed, not documents, requests, or words.
Question 10
Your company stores hundreds of internal business reports.
You need to recommend a generative AI solution that uses an agent to answer questions based on the content in the reports.
What should you include in the recommendation?
A. natural language processing
B. Azure Machine Learning
C. Retrieval Augmented Generation (RAG)
D. Generative Adversarial Network (GAN)
Show Answer
Correct Answer: C
Explanation: Retrieval Augmented Generation (RAG) is the standard approach for building generative AI agents that answer questions using an organization's internal documents. It retrieves relevant report content and grounds the model's responses in that retrieved context, avoiding the need to retrain the model. Natural language processing is too broad, Azure Machine Learning is a platform rather than the specific solution pattern, and GANs are used for data generation, not document question answering.
Question 11
Your company uses generative AI to assist with content creation and customer interactions.
You need to evaluate whether Azure Machine Learning can add value to the current customer management.
For which use case should you use Machine Learning?
A. predicting customer retention
B. creating product descriptions from images
C. generating marketing campaigns
D. summarizing customer service transcripts
Show Answer
Correct Answer: A
Explanation: Azure Machine Learning is best suited for predictive analytics using historical data. Predicting customer retention (or churn) is a classic supervised machine learning use case. The other options—creating product descriptions from images, generating marketing campaigns, and summarizing customer service transcripts—are primarily generative AI tasks typically handled by large language or multimodal models rather than traditional Azure Machine Learning predictive modeling.
$19
Get all 99 questions with detailed answers and explanations
Instant download HTML + PDF delivered the moment payment clears.
Secure Stripe checkout we never see or store your card details.
7-day refund if files are defective see our refund policy.
