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 55

Question 540

An Azure-based AI team must satisfy this objective: A recruitment assistant scores applicants differently when identical qualifications are paired with different names. Which design choice is best?

  1. Use validated data retrieval and source-backed numerical responses
  2. Evaluate outcomes with counterfactual name swaps and remove unjustified proxy effects
  3. Record the actual decision output, input reference, model and prompt versions in integrity-protected audit records
  4. Test representative document layouts and accessibility-oriented inputs

Explanation

Correct Answer

B:

Fairness checks compare materially equivalent cases rather than raw group averages alone. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

A numeric-looking answer is not proof that a measure was computed. That is not sufficient here: Fairness checks compare materially equivalent cases rather than raw group averages alone.

C:

Auditability requires the original decision evidence and version identifiers; regenerating model text is not guaranteed to be identical. That is not sufficient here: Fairness checks compare materially equivalent cases rather than raw group averages alone.

D:

Inclusion requires robust input handling beyond a single document template. That is not sufficient here: Fairness checks compare materially equivalent cases rather than raw group averages alone.

Question 541

Troubleshooting report: The SDK receives raw bytes without the format expected by its stream. Which correction is MOST directly targeted to this failure?

  1. Use explicit language confirmation or a support fallback
  2. Close the capture stream and update permission controls
  3. Use the correct audio encoding and recognized input format
  4. Apply recognized punctuation support and validate the rendered transcript

Explanation

Correct Answer

C:

This correction directly addresses the reported failure. Speech recognition requires the service to decode audio correctly. The targeted change is: Use the correct audio encoding and recognized input format.

Incorrect Answers

A:

Unrecognized languages must not be silently interpreted as a known one. In the stated case, troubleshooting report: The SDK receives raw bytes without the format expected by its stream; “Use explicit language confirmation or a support fallback” does not target that failure.

B:

User control of capture must be enforced at the device or application layer. In the stated case, troubleshooting report: The SDK receives raw bytes without the format expected by its stream; “Close the capture stream and update permission controls” does not target that failure.

D:

Speech text may require punctuation inference distinct from word recognition. In the stated case, troubleshooting report: The SDK receives raw bytes without the format expected by its stream; “Apply recognized punctuation support and validate the rendered transcript” does not target that failure.

Question 542

An Azure AI Fundamentals solution has this requirement: A claims assistant includes full customer identifiers in diagnostic traces. Which implementation most directly addresses it?

  1. Provide semantic labels and keyboard-operable controls for all important interactions
  2. Mark synthetic examples clearly and require verified consent for attributed statements
  3. Maintain a governed feature inventory and trace decisions to policy
  4. Redact personal fields before telemetry storage and control log access

Explanation

Correct Answer

D:

Troubleshooting does not justify reproducing sensitive application records. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Inclusive design permits nonvisual navigation of the same task path. That is not sufficient here: Troubleshooting does not justify reproducing sensitive application records.

B:

Transparency prohibits presenting invented testimony as genuine. That is not sufficient here: Troubleshooting does not justify reproducing sensitive application records.

C:

Accountability requires interpretable reasons for consequential decisions. That is not sufficient here: Troubleshooting does not justify reproducing sensitive application records.

Question 543

The approved design action is: Use an image-generation-capable model with an appropriate prompt. Which test result would BEST establish that it worked?

  1. A returned serial number is stored as OCR text not a generic image tag
  2. The output is a new raster image matching the approved icon brief
  3. Selected candidates differ in composition while meeting required constraints
  4. The answer reproduces the visible sign text accurately

Explanation

Correct Answer

B:

Image generation creates new pixels rather than classifying an existing photo. The expected acceptance evidence is: The output is a new raster image matching the approved icon brief. This observation directly tests the stated behavior.

Incorrect Answers

A:

Tags describe concepts while OCR reports character content. This observation does not establish the required result: The output is a new raster image matching the approved icon brief.

C:

Image-generation variability can be useful but does not guarantee suitability. This observation does not establish the required result: The output is a new raster image matching the approved icon brief.

D:

Visual evidence is needed to read sign content. This observation does not establish the required result: The output is a new raster image matching the approved icon brief.

Question 544

A project needs the following AI behavior: A helpdesk receives written feedback and wants a positive or negative sentiment label. What should the designer implement?

  1. Recognize speech, translate text, then synthesize French output
  2. Use text summarization with clear length and fidelity constraints
  3. Use word or utterance timestamps from a supported recognition output
  4. Call a supported text sentiment-analysis feature and interpret returned polarity

Explanation

Correct Answer

D:

Sentiment output classifies attitudes expressed in the submitted text. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Cross-language audio output requires both language transfer and synthesis. That is not sufficient here: Sentiment output classifies attitudes expressed in the submitted text.

B:

Summarization compresses written information rather than changing its facts. That is not sufficient here: Sentiment output classifies attitudes expressed in the submitted text.

C:

Timecodes align text segments to audio playback. That is not sufficient here: Sentiment output classifies attitudes expressed in the submitted text.

Question 545

The approved design action is: Send the audio through a deployment that supports audio inputs. Which test result would BEST establish that it worked?

  1. The model's response reflects the words spoken in the actual recording
  2. Subtitle cues match the spoken phrases at measured times
  3. The narration pauses at marked section boundaries
  4. The spoken answer slows while preserving its content

Explanation

Correct Answer

A:

A text-only request cannot interpret sound content by file name alone. The expected acceptance evidence is: The model's response reflects the words spoken in the actual recording. This observation directly tests the stated behavior.

Incorrect Answers

B:

Timecodes align text segments to audio playback. This observation does not establish the required result: The model's response reflects the words spoken in the actual recording.

C:

SSML controls aspects of prosody and timing for synthesized speech. This observation does not establish the required result: The model's response reflects the words spoken in the actual recording.

D:

Voice selection and prosody can influence delivery without changing the text. This observation does not establish the required result: The model's response reflects the words spoken in the actual recording.

Question 546

An Azure-based AI team must satisfy this objective: A user asks to list how many identical packages appear in a photograph. Which design choice is best?

  1. Apply bounded retries and respect the service's throughput limit
  2. Choose detection only if spatial object locations are required
  3. Detect individual package instances and count nonduplicate boxes
  4. Request image tagging for the selected visual features

Explanation

Correct Answer

C:

Object counts require resolving distinct instances, not trusting one scene label. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Too many concurrent requests can exceed resource quota. That is not sufficient here: Object counts require resolving distinct instances, not trusting one scene label.

B:

Classification predicts image-level categories while detection provides positions. That is not sufficient here: Object counts require resolving distinct instances, not trusting one scene label.

D:

Tags summarize recognized concepts as labels. That is not sufficient here: Object counts require resolving distinct instances, not trusting one scene label.

Question 547

An Azure AI solution uses the following approved control: Use document extraction with source positions before indexing for search. What observable result should pass QA?

  1. Observed labeled events receive evidence-backed intervals and brief unobserved events are flagged for review
  2. A question links to the correct contract section and page
  3. Each known document type reaches the expected analyzer
  4. The answer resolves to a specific approved source document version

Explanation

Correct Answer

B:

Structured layout and provenance let users verify a quoted clause. The expected acceptance evidence is: A question links to the correct contract section and page. This observation directly tests the stated behavior.

Incorrect Answers

A:

Temporal evidence supports the timing of observed events, but approximately one frame per second cannot guarantee subsecond event capture. This observation does not establish the required result: A question links to the correct contract section and page.

C:

Different document kinds have distinct field schemas and review paths. This observation does not establish the required result: A question links to the correct contract section and page.

D:

Verifiable provenance differentiates a cited answer from guesswork. This observation does not establish the required result: A question links to the correct contract section and page.

Question 548

An engineering review documents this requirement: A team needs to compare two speech-recognition configurations. Select the most appropriate solution.

  1. Evaluate word error rate on the same held-out audio samples
  2. Use word or utterance timestamps from a supported recognition output
  3. Use a supported text-to-speech synthesis deployment
  4. Match the speech resource region or endpoint to the SDK configuration

Explanation

Correct Answer

A:

Comparable labeled audio prevents misleading quality claims. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

B:

Timecodes align text segments to audio playback. That is not sufficient here: Comparable labeled audio prevents misleading quality claims.

C:

Speech synthesis generates audio waveforms from supplied text. That is not sufficient here: Comparable labeled audio prevents misleading quality claims.

D:

Region or endpoint mismatch prevents requests reaching the correct resource. That is not sufficient here: Comparable labeled audio prevents misleading quality claims.

Question 549

An Azure-based AI team must satisfy this objective: A voice bot tries to synthesize a sentence containing special XML characters in SSML. Which design choice is best?

  1. Start the stream and recognition lifecycle before discarding buffered speech
  2. Escape XML special characters and validate SSML syntax
  3. Configure a supported Spanish recognition locale and test representative speakers
  4. Select a supported synthesis voice ID rather than altering transcript text

Explanation

Correct Answer

B:

SSML is XML and malformed markup breaks the synthesis request. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Real-time audio pipelines need coordinated capture and recognition start. That is not sufficient here: SSML is XML and malformed markup breaks the synthesis request.

C:

Recognition models need the appropriate supported language settings. That is not sufficient here: SSML is XML and malformed markup breaks the synthesis request.

D:

Voice timbre is a synthesis configuration, not speech recognition output. That is not sufficient here: SSML is XML and malformed markup breaks the synthesis request.

Question 550

An AI operations team is testing two distinct capabilities. Failure (1): The checkbox is discarded during plain text extraction. Failure (2): The model fabricates a purchase-order number. Which TWO specific corrections should the team apply?

  1. Keep transcript anchors and validate extracted tasks
  2. Include selection-mark processing in the analyzer
  3. Use table-aware layout extraction and numeric validation
  4. Enable supported layout and field extraction features
  5. Validate field presence against source evidence

Explanation

Correct Answers

B:

Correct for failure (1). Checkbox state is a structured signal rather than arbitrary prose. The proposed correction directly addresses that reported defect.

E:

Correct for failure (2). Extraction should preserve absence and distinguish it from zero. It independently corrects the second defect.

Incorrect Answers

A:

Transcription and summarization are separate results with different validation. The stated pair of failures instead requires “Include selection-mark processing in the analyzer” and “Validate field presence against source evidence”; this option cannot substitute for both.

C:

Cell relationships are necessary to check item totals. It does not complete the two decisions required in this question.

D:

Ordinary plain-text OCR may lose table relationships and checkbox states. The stated pair of failures instead requires “Include selection-mark processing in the analyzer” and “Validate field presence against source evidence”; this option cannot substitute for both.

Microsoft Azure AI Fundamentals (AI-901) Practice Tests