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 81

Question 800

Troubleshooting report: The retrieval system ranks an expired document highest by semantic similarity. Which correction is MOST directly targeted to this failure?

  1. Add adversarial and exception-focused cases before promotion
  2. Raise high-risk routing and restrict unsupported actionable advice
  3. Filter sources by effective status before similarity ranking
  4. Exclude obsolete material with approval and timestamp filters

Explanation

Correct Answer

C:

This correction directly addresses the reported failure. Truthfulness requires evidence that corresponds to the operative version. The targeted change is: Filter sources by effective status before similarity ranking.

Incorrect Answers

A:

A high overall score can conceal severe failure on unusual inputs. In the stated case, troubleshooting report: The retrieval system ranks an expired document highest by semantic similarity; “Add adversarial and exception-focused cases before promotion” does not target that failure.

B:

Safety mechanisms should reduce unsupported high-impact recommendations. In the stated case, troubleshooting report: The retrieval system ranks an expired document highest by semantic similarity; “Raise high-risk routing and restrict unsupported actionable advice” does not target that failure.

D:

A precise explanation of an obsolete procedure can still be dangerous. In the stated case, troubleshooting report: The retrieval system ranks an expired document highest by semantic similarity; “Exclude obsolete material with approval and timestamp filters” does not target that failure.

Question 801

An acceptance tester needs to validate: Limit context to necessary messages and apply safe summaries or retrieval. Which observed result proves the intended behavior?

  1. A sample call passes schema validation and succeeds
  2. An authenticated test request returns a response from the chosen deployment
  3. The same task passes with reduced token use and no quality loss
  4. The second tenant's test passes with its own authorized identity

Explanation

Correct Answer

C:

Long irrelevant context consumes tokens and may increase latency. The expected acceptance evidence is: The same task passes with reduced token use and no quality loss. This observation directly tests the stated behavior.

Incorrect Answers

A:

Invalid tool parameters break calls even if the intended operation is correct. This observation does not establish the required result: The same task passes with reduced token use and no quality loss.

B:

The portal provides model selection, deployment and test interfaces for supported resources. This observation does not establish the required result: The same task passes with reduced token use and no quality loss.

D:

Copying code does not grant access to another tenant's resources. This observation does not establish the required result: The same task passes with reduced token use and no quality loss.

Question 802

An engineering review documents this requirement: An AI agent receives a web page that instructs it to ignore prior safety policies. Select the most appropriate solution.

  1. Verify permissions through the identity provider and server-side roles
  2. Require governance evidence with reviewers, test outcomes, and documented exceptions
  3. Treat retrieved content as untrusted evidence and prevent it from changing tool authority
  4. Identify the virtual agent as synthetic and avoid deceptive identity cues

Explanation

Correct Answer

C:

Lower-trust text is not a privileged command channel. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Natural-language claims cannot elevate authentication authority. That is not sufficient here: Lower-trust text is not a privileged command channel.

B:

Self-attestation is not independent proof of regulatory controls. That is not sufficient here: Lower-trust text is not a privileged command channel.

D:

Transparency prevents users from falsely assuming they are speaking with a human. That is not sufficient here: Lower-trust text is not a privileged command channel.

Question 803

The approved design action is: Use invoice-oriented structured extraction with table row preservation. Which test result would BEST establish that it worked?

  1. The minutes correctly associate tasks with speakers and source timestamps
  2. The timeline marks arrival frames with appropriate timestamps
  3. Line-item totals and row relationships reconcile to the document
  4. The product shape remains unchanged against the new background

Explanation

Correct Answer

C:

Table extraction associates quantities, descriptions and prices by row. The expected acceptance evidence is: Line-item totals and row relationships reconcile to the document. This observation directly tests the stated behavior.

Incorrect Answers

A:

Attribution distinguishes who committed to each action. This observation does not establish the required result: Line-item totals and row relationships reconcile to the document.

B:

Video understanding retains order and time, unlike single-image tagging. This observation does not establish the required result: Line-item totals and row relationships reconcile to the document.

D:

Editing changes selected pixels while leaving product regions intact. This observation does not establish the required result: Line-item totals and row relationships reconcile to the document.

Question 804

An Azure-based AI team must satisfy this objective: An expense review system must read fields from a photographed receipt. Which design choice is best?

  1. Use visual document extraction with appropriate field and confidence rules
  2. Use text summarization with fidelity checks against sources
  3. Use object detection and count accepted object instances
  4. Use OCR tuned to the source quality and validate extracted text

Explanation

Correct Answer

A:

Images of receipts need OCR/layout extraction before typed business values. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

B:

Summarization compresses content without independently validating missing facts. That is not sufficient here: Images of receipts need OCR/layout extraction before typed business values.

C:

Instance detections support counts where one image contains several objects. That is not sufficient here: Images of receipts need OCR/layout extraction before typed business values.

D:

Text recognition reads characters rather than merely describing visual objects. That is not sufficient here: Images of receipts need OCR/layout extraction before typed business values.

Question 805

An AI operations team is testing two distinct capabilities. Failure (1): Shared cached context is reused across unrelated accounts. Failure (2): The system adds fabricated citations when search returns no documents. Which TWO specific corrections should the team apply?

  1. Require evidence-backed citation construction and abstention on no results
  2. Partition state by user and clear or secure session context
  3. Remove the false attribution and require documentary consent
  4. Add adversarial and exception-focused cases before promotion
  5. Assess both unsafe misses and harmful overblocking

Explanation

Correct Answers

A:

Correct for failure (2). Honesty requires acknowledging when a claim cannot be substantiated. It independently corrects the second defect.

B:

Correct for failure (1). Privacy includes preventing cross-session memory disclosure. The proposed correction directly addresses that reported defect.

Incorrect Answers

C:

Transparency prohibits presenting invented testimony as genuine. It does not complete the two decisions required in this question.

D:

A high overall score can conceal severe failure on unusual inputs. The stated pair of failures instead requires “Require evidence-backed citation construction and abstention on no results” and “Partition state by user and clear or secure session context”; this option cannot substitute for both.

E:

Safety controls should not overblock permitted assistance. It does not complete the two decisions required in this question.

Question 806

An Azure AI solution uses the following approved control: Handle tool timeout as a typed error and preserve partial workflow state. What observable result should pass QA?

  1. A simulated timeout yields pending status with a safe retry or escalation
  2. Test requests consistently produce parseable objects with required keys
  3. A trace shows which tool result supported each claimed business action
  4. An echo-free test response correctly addresses the supplied question

Explanation

Correct Answer

A:

Timeout handling should not be reported as a successful action. The expected acceptance evidence is: A simulated timeout yields pending status with a safe retry or escalation. This observation directly tests the stated behavior.

Incorrect Answers

B:

Explicit examples can improve adherence to bounded output conventions. This observation does not establish the required result: A simulated timeout yields pending status with a safe retry or escalation.

C:

Operational observability aids debugging without exposing hidden sensitive prompt data. This observation does not establish the required result: A simulated timeout yields pending status with a safe retry or escalation.

D:

The inference service needs task content to generate a relevant answer. This observation does not establish the required result: A simulated timeout yields pending status with a safe retry or escalation.

Question 807

A solution team has selected this control: Evaluate instructions and fallback quality in each supported language. Which observation is the most direct acceptance evidence?

  1. An unprivileged test user is denied protected operations despite claiming to be an admin
  2. Language-specific test cases produce equivalent safe outcomes and escalation options
  3. Exceptional high-impact cases are tested with explicit pass criteria
  4. Accent-specific word error rates meet stated thresholds or trigger alternatives

Explanation

Correct Answer

B:

Inclusion requires equivalent safety and meaning across supported locales. The expected acceptance evidence is: Language-specific test cases produce equivalent safe outcomes and escalation options. This observation directly tests the stated behavior.

Incorrect Answers

A:

Natural-language claims cannot elevate authentication authority. This observation does not establish the required result: Language-specific test cases produce equivalent safe outcomes and escalation options.

C:

A high overall score can conceal severe failure on unusual inputs. This observation does not establish the required result: Language-specific test cases produce equivalent safe outcomes and escalation options.

D:

Inclusive voice systems require performance evaluation beyond a single accent. This observation does not establish the required result: Language-specific test cases produce equivalent safe outcomes and escalation options.

Question 808

Troubleshooting report: Individual users can edit the system behavior stored in their own messages. Which correction is MOST directly targeted to this failure?

  1. Move system instructions into protected application configuration
  2. Retain bounded conversation state with relevant history
  3. Update configuration and run supported compatibility checks
  4. Apply case and role filters before providing model context

Explanation

Correct Answer

A:

This correction directly addresses the reported failure. Versioned system messages provide a controlled behavioral baseline. The targeted change is: Move system instructions into protected application configuration.

Incorrect Answers

B:

Stateful conversations require supplying or retrieving relevant prior context. In the stated case, troubleshooting report: Individual users can edit the system behavior stored in their own messages; “Retain bounded conversation state with relevant history” does not target that failure.

C:

A model family label is not necessarily the concrete deployed endpoint. In the stated case, troubleshooting report: Individual users can edit the system behavior stored in their own messages; “Update configuration and run supported compatibility checks” does not target that failure.

D:

Source scoping enforces privacy and also improves answer relevance. In the stated case, troubleshooting report: Individual users can edit the system behavior stored in their own messages; “Apply case and role filters before providing model context” does not target that failure.

Question 809

An engineering review documents this requirement: A user requests deletion of personal interaction data kept for agent memory. Select the most appropriate solution.

  1. Identify the virtual agent as synthetic and avoid deceptive identity cues
  2. Restrict retrieval to effective approved documents and expose source dates
  3. Provide semantic labels and keyboard-operable controls for all important interactions
  4. Apply retention and deletion controls subject to recorded legal exceptions

Explanation

Correct Answer

D:

Accountability requires evidence of actual data lifecycle actions. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Transparency prevents users from falsely assuming they are speaking with a human. That is not sufficient here: Accountability requires evidence of actual data lifecycle actions.

B:

Truthfulness requires evidence that corresponds to the operative version. That is not sufficient here: Accountability requires evidence of actual data lifecycle actions.

C:

Inclusive design permits nonvisual navigation of the same task path. That is not sufficient here: Accountability requires evidence of actual data lifecycle actions.

Question 810

An AI operations team is testing two distinct capabilities. Failure (1): The summarizer uses only a single thumbnail to infer all events. Failure (2): The search index exposes all extracted pages to every client. Which TWO specific corrections should the team apply?

  1. Enforce source permissions in indexing and query-time filtering
  2. Represent repeated fields as arrays and validate counts
  3. Inspect multiple temporal segments and independently review brief events that sampled frames could miss
  4. Build citations from stored IDs and source spans
  5. Keep transcript anchors and validate extracted tasks

Explanation

Correct Answers

A:

Correct for failure (2). Extracting content does not waive document access controls. It independently corrects the second defect.

C:

Correct for failure (1). Content Understanding samples approximately one frame per second, so brief or single-frame actions can be missed despite valid timestamps. The proposed correction directly addresses that reported defect.

Incorrect Answers

B:

Repeated records require structured lists with item-level associations. It does not complete the two decisions required in this question.

D:

Verifiable provenance differentiates a cited answer from guesswork. It does not complete the two decisions required in this question.

E:

Transcription and summarization are separate results with different validation. The stated pair of failures instead requires “Enforce source permissions in indexing and query-time filtering” and “Inspect multiple temporal segments and independently review brief events that sampled frames could miss”; this option cannot substitute for both.

Microsoft Azure AI Fundamentals (AI-901) Practice Tests