McAfee Secure

Exam Code: AI-901

Exam Name: Microsoft Azure AI Fundamentals

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Azure AI Fundamentals

Microsoft AI-901 Questions & Answers

Study with Up-To-Date REAL Exam Questions and Answers from the ACTUAL Test

166 Questions & Answers with Testing Engine

"Microsoft Azure AI Fundamentals Exam", also known as AI-901 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date AI-901 Exam Engine. Your AI-901 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft AI-901 Practice Test 71

Question 700

Troubleshooting report: The prototype launches unnecessary specialist agents for one lookup. Which correction is MOST directly targeted to this failure?

  1. Move system instructions into protected application configuration
  2. Exercise negative and expected tool paths before release
  3. Apply case and role filters before providing model context
  4. Create and test a simple one-agent tool workflow

Explanation

Correct Answer

D:

This correction directly addresses the reported failure. A single agent can orchestrate a bounded tool without a multi-agent system. The targeted change is: Create and test a simple one-agent tool workflow.

Incorrect Answers

A:

Versioned system messages provide a controlled behavioral baseline. In the stated case, troubleshooting report: The prototype launches unnecessary specialist agents for one lookup; “Move system instructions into protected application configuration” does not target that failure.

B:

An answer that sounds plausible may conceal failed tool calls. In the stated case, troubleshooting report: The prototype launches unnecessary specialist agents for one lookup; “Exercise negative and expected tool paths before release” does not target that failure.

C:

Source scoping enforces privacy and also improves answer relevance. In the stated case, troubleshooting report: The prototype launches unnecessary specialist agents for one lookup; “Apply case and role filters before providing model context” does not target that failure.

Question 701

An Azure-based AI team must satisfy this objective: A retailer wants to group shoppers by purchasing similarity without labeled categories. Which design choice is best?

  1. Use supervised classification with labeled maintenance outcomes
  2. Use a generative language workload with policy-grounded response constraints
  3. Use invoice-oriented structured extraction with table row preservation
  4. Use clustering with normalized behavioral features

Explanation

Correct Answer

D:

Clustering discovers groups based on feature similarity. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

A failure probability or class is a discrete decision under a defined horizon. That is not sufficient here: Clustering discovers groups based on feature similarity.

B:

Generation produces novel text, unlike choosing a fixed class label. That is not sufficient here: Clustering discovers groups based on feature similarity.

C:

Table extraction associates quantities, descriptions and prices by row. That is not sufficient here: Clustering discovers groups based on feature similarity.

Question 702

An acceptance tester needs to validate: Place the current request in the supported user message field. Which observed result proves the intended behavior?

  1. Questions without a matching manual passage yield a qualified response
  2. The second tenant's test passes with its own authorized identity
  3. An echo-free test response correctly addresses the supplied question
  4. A trace shows the task finished and its output is available

Explanation

Correct Answer

C:

The inference service needs task content to generate a relevant answer. The expected acceptance evidence is: An echo-free test response correctly addresses the supplied question. This observation directly tests the stated behavior.

Incorrect Answers

A:

Retrieval access and abstention avoid inventing unsupported documentation. This observation does not establish the required result: An echo-free test response correctly addresses the supplied question.

B:

Copying code does not grant access to another tenant's resources. This observation does not establish the required result: An echo-free test response correctly addresses the supplied question.

D:

Submission is not the same as successful completion. This observation does not establish the required result: An echo-free test response correctly addresses the supplied question.

Question 703

An Azure AI solution uses the following approved control: Maintain bounded authenticated conversation state with order context. What observable result should pass QA?

  1. An echo-free test response correctly addresses the supplied question
  2. The same authorized case ID is observed in every step receipt
  3. A follow-up refers to the correct permitted order from the same session
  4. An out-of-scope request is handled under the approved restrictions

Explanation

Correct Answer

C:

Conversation continuity should preserve identifiers without crossing user sessions. The expected acceptance evidence is: A follow-up refers to the correct permitted order from the same session. This observation directly tests the stated behavior.

Incorrect Answers

A:

The inference service needs task content to generate a relevant answer. This observation does not establish the required result: A follow-up refers to the correct permitted order from the same session.

B:

Structured state prevents subtle substitution of unrelated customer records. This observation does not establish the required result: A follow-up refers to the correct permitted order from the same session.

D:

High-priority instructions should describe approved behavior and escalation rules. This observation does not establish the required result: A follow-up refers to the correct permitted order from the same session.

Question 704

An Azure AI solution uses the following approved control: Require independent transaction authorization and approval outside prompts. What observable result should pass QA?

  1. The transfer remains pending until delegated approval is recorded
  2. Each policy answer citation resolves to the correct approved paragraph
  3. Above-limit test refunds remain pending until an authorized approver acts
  4. Eligible memory entries disappear from active stores after the verified request

Explanation

Correct Answer

A:

Accountable high-impact operations need verifiable authorized signatures. The expected acceptance evidence is: The transfer remains pending until delegated approval is recorded. This observation directly tests the stated behavior.

Incorrect Answers

B:

Transparency lets users inspect evidence instead of accepting fluent text. This observation does not establish the required result: The transfer remains pending until delegated approval is recorded.

C:

A prompt-level request cannot override a transaction policy. This observation does not establish the required result: The transfer remains pending until delegated approval is recorded.

D:

Accountability requires evidence of actual data lifecycle actions. This observation does not establish the required result: The transfer remains pending until delegated approval is recorded.

Question 705

Troubleshooting report: The extractor chooses the first currency value in the contract. Which correction is MOST directly targeted to this failure?

  1. Use schema definitions and grounding evidence for each field
  2. Preserve conflict flags and apply approved precedence rules
  3. Use typed currency extraction and explicit ambiguity handling
  4. Enable document segmentation and per-document extraction

Explanation

Correct Answer

A:

This correction directly addresses the reported failure. Custom extraction maps semantically relevant phrases to typed fields. The targeted change is: Use schema definitions and grounding evidence for each field.

Incorrect Answers

B:

A model should not silently choose among competing contractual evidence. In the stated case, troubleshooting report: The extractor chooses the first currency value in the contract; “Preserve conflict flags and apply approved precedence rules” does not target that failure.

C:

A currency symbol or amount alone may not establish financial meaning. In the stated case, troubleshooting report: The extractor chooses the first currency value in the contract; “Use typed currency extraction and explicit ambiguity handling” does not target that failure.

D:

Document boundaries prevent values from unrelated invoices being joined. In the stated case, troubleshooting report: The extractor chooses the first currency value in the contract; “Enable document segmentation and per-document extraction” does not target that failure.

Question 706

An Azure-based AI team must satisfy this objective: A retail product photo contains text, a logo, and a person. Which design choice is best?

  1. Keep recognized visual labels separate from text-read lines and words
  2. Select appropriate separate visual features for OCR, tagging and people detection
  3. Generate and edit concise factual alt text describing significant content
  4. Expose uncertainty and route low-confidence image decisions to review

Explanation

Correct Answer

B:

Different visual outputs describe text, scene classes and detected people. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Tags describe concepts while OCR reports character content. That is not sufficient here: Different visual outputs describe text, scene classes and detected people.

C:

Accessible captions should communicate image purpose without invented details. That is not sufficient here: Different visual outputs describe text, scene classes and detected people.

D:

Generated visual interpretations can be wrong even when phrased fluently. That is not sufficient here: Different visual outputs describe text, scene classes and detected people.

Question 707

An AI operations team is testing two distinct capabilities. Failure (1): The public JavaScript bundle contains the service API key. Failure (2): The agent relies on free-form generation when retrieval is empty. Which TWO specific corrections should the team apply?

  1. Tighten the output contract and apply post-response validation
  2. Enforce source filters and explicit missing-evidence handling
  3. Respect retry guidance and cap concurrent calls
  4. Move the model call behind a secured server endpoint
  5. Freeze representative evaluation cases and compare versions

Explanation

Correct Answers

B:

Correct for failure (2). Retrieval access and abstention avoid inventing unsupported documentation. It independently corrects the second defect.

D:

Correct for failure (1). Backend controls protect credentials, quotas and data policy. The proposed correction directly addresses that reported defect.

Incorrect Answers

A:

Explicit examples can improve adherence to bounded output conventions. The stated pair of failures instead requires “Enforce source filters and explicit missing-evidence handling” and “Move the model call behind a secured server endpoint”; this option cannot substitute for both.

C:

Rate limits protect the service and uncontrolled retries amplify overload. The stated pair of failures instead requires “Enforce source filters and explicit missing-evidence handling” and “Move the model call behind a secured server endpoint”; this option cannot substitute for both.

E:

Anecdotal success on different prompts cannot establish an improvement. The stated pair of failures instead requires “Enforce source filters and explicit missing-evidence handling” and “Move the model call behind a secured server endpoint”; this option cannot substitute for both.

Question 708

An engineering review documents this requirement: A support system wants to summarize chats but must avoid processing unrelated private threads. Select the most appropriate solution.

  1. Limit retrieval and chat context to the authenticated case scope
  2. Archive the actual evaluated response together with the prompt, model deployment, input and generation settings
  3. Check supported context and output limits for the chosen deployment
  4. Choose a model size using representative quality and cost benchmarks

Explanation

Correct Answer

A:

Source scoping enforces privacy and also improves answer relevance. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

B:

Recorded settings support investigation, but generative models can still vary on an apparently identical rerun. That is not sufficient here: Source scoping enforces privacy and also improves answer relevance.

C:

Capabilities and limits vary by model deployment. That is not sufficient here: Source scoping enforces privacy and also improves answer relevance.

D:

Lower-cost deployment is acceptable only if accuracy and safety floors remain met. That is not sufficient here: Source scoping enforces privacy and also improves answer relevance.

Question 709

The approved design action is: Provide synchronized transcripts or captions for spoken content. Which test result would BEST establish that it worked?

  1. Accent-specific word error rates meet stated thresholds or trigger alternatives
  2. A first-use test displays clear AI identification and human contact option
  3. A deaf tester receives the complete equivalent content as text
  4. An unprivileged test user is denied protected operations despite claiming to be an admin

Explanation

Correct Answer

C:

Accessible information must not depend on hearing audio. The expected acceptance evidence is: A deaf tester receives the complete equivalent content as text. This observation directly tests the stated behavior.

Incorrect Answers

A:

Inclusive voice systems require performance evaluation beyond a single accent. This observation does not establish the required result: A deaf tester receives the complete equivalent content as text.

B:

Transparency helps users assess the scope and limitations of the assistance. This observation does not establish the required result: A deaf tester receives the complete equivalent content as text.

D:

Natural-language claims cannot elevate authentication authority. This observation does not establish the required result: A deaf tester receives the complete equivalent content as text.

Question 710

An Azure AI Fundamentals solution has this requirement: A client needs to track whether a long-running agent task has completed. Which implementation most directly addresses it?

  1. Preserve the higher-priority application constraints in the message hierarchy
  2. Align the tool argument names and types with the published function schema
  3. Use the supported run or response status and verify its terminal state
  4. Use a compatible code-capable generative model with constrained context and tests

Explanation

Correct Answer

C:

Submission is not the same as successful completion. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Developer and system constraints govern allowed operations over user requests. That is not sufficient here: Submission is not the same as successful completion.

B:

Invalid tool parameters break calls even if the intended operation is correct. That is not sufficient here: Submission is not the same as successful completion.

D:

Generated code must be validated, not treated as production-ready from fluency. That is not sufficient here: Submission is not the same as successful completion.

Microsoft Azure AI Fundamentals (AI-901) Practice Tests