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 11

Question 100

An acceptance tester needs to validate: Enable supported document-field confidence and grounding, then calibrate field-specific review thresholds using labeled samples. Which observed result proves the intended behavior?

  1. Mixed-language fields are transcribed correctly on labeled samples
  2. A reviewer confirms the selected clauses and visible signature region from the source, without claiming a verified legal signature
  3. Validated fields above tested thresholds proceed while uncertain or inconsistent fields enter review
  4. Extracted rows align with the original form cells

Explanation

Correct Answer

C:

Content Understanding confidence applies to supported document analyzers; confidence estimates alone do not prove field correctness. The expected acceptance evidence is: Validated fields above tested thresholds proceed while uncertain or inconsistent fields enter review. This observation directly tests the stated behavior.

Incorrect Answers

A:

Recognition accuracy depends on script and language support. This observation does not establish the required result: Validated fields above tested thresholds proceed while uncertain or inconsistent fields enter review.

B:

Selection-mark extraction is available with GA document analysis, but dedicated signature-location detection is available only in the 2026-06-01-preview API and cannot establish legal validity. This observation does not establish the required result: Validated fields above tested thresholds proceed while uncertain or inconsistent fields enter review.

D:

Ordinary plain-text OCR may lose table relationships and checkbox states. This observation does not establish the required result: Validated fields above tested thresholds proceed while uncertain or inconsistent fields enter review.

Question 101

Troubleshooting report: The model answers from the text question without receiving the screenshot. Which correction is MOST directly targeted to this failure?

  1. Use key phrase extraction rather than regression
  2. Provide the supported image input alongside the question
  3. Call the authorized ticket tool and verify the receipt
  4. Define extraction fields and validate values

Explanation

Correct Answer

B:

This correction directly addresses the reported failure. Image-and-text reasoning can connect a question with visible scene content. The targeted change is: Provide the supported image input alongside the question.

Incorrect Answers

A:

Key phrase extraction finds prominent topics without inventing new source material. In the stated case, troubleshooting report: The model answers from the text question without receiving the screenshot; “Use key phrase extraction rather than regression” does not target that failure.

C:

Agentic workloads combine reasoning and bounded external actions. In the stated case, troubleshooting report: The model answers from the text question without receiving the screenshot; “Call the authorized ticket tool and verify the receipt” does not target that failure.

D:

Document extraction maps unstructured layouts into structured values. In the stated case, troubleshooting report: The model answers from the text question without receiving the screenshot; “Define extraction fields and validate values” does not target that failure.

Question 102

An Azure-based AI team must satisfy this objective: An agent should answer a spoken question directly using a multimodal model. Which design choice is best?

  1. Verify the service credential, resource and authorized region
  2. Send the audio through a deployment that supports audio inputs
  3. Apply approved audio retention and access rules and keep only required derived text
  4. Match the speech resource region or endpoint to the SDK configuration

Explanation

Correct Answer

B:

A text-only request cannot interpret sound content by file name alone. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Audio capture success does not establish remote service authentication. That is not sufficient here: A text-only request cannot interpret sound content by file name alone.

C:

Speech workloads still process personal data and require controlled lifecycle. That is not sufficient here: A text-only request cannot interpret sound content by file name alone.

D:

Region or endpoint mismatch prevents requests reaching the correct resource. That is not sufficient here: A text-only request cannot interpret sound content by file name alone.

Question 103

A solution team has selected this control: Evaluate noise conditions and input quality rather than only clean studio speech. Which observation is the most direct acceptance evidence?

  1. The recognizer succeeds with valid resource credentials
  2. Noise-augmented recognition tests meet requirements or request repetition
  3. The selected voice reads the intended language correctly
  4. Labeled sample complaints and praise yield expected sentiment categories

Explanation

Correct Answer

B:

Signal quality directly affects recognizer error rates. The expected acceptance evidence is: Noise-augmented recognition tests meet requirements or request repetition. This observation directly tests the stated behavior.

Incorrect Answers

A:

Audio capture success does not establish remote service authentication. This observation does not establish the required result: Noise-augmented recognition tests meet requirements or request repetition.

C:

Not every voice is supported for all languages. This observation does not establish the required result: Noise-augmented recognition tests meet requirements or request repetition.

D:

Sentiment output classifies attitudes expressed in the submitted text. This observation does not establish the required result: Noise-augmented recognition tests meet requirements or request repetition.

Question 104

The approved design action is: Separate saving, testing and publication actions with explicit permission. Which test result would BEST establish that it worked?

  1. The test output satisfies every defined requirement of the prompt
  2. The service processes role-tagged messages with the expected application behavior
  3. An authorized release control is required before user-facing access changes
  4. Execution traces show the intended tool invoked with valid arguments

Explanation

Correct Answer

C:

A draft or tested agent is not automatically a publicly released service. The expected acceptance evidence is: An authorized release control is required before user-facing access changes. This observation directly tests the stated behavior.

Incorrect Answers

A:

Specific instructions reduce ambiguity about what constitutes success. This observation does not establish the required result: An authorized release control is required before user-facing access changes.

B:

Chat APIs use structured messages rather than concatenating untyped command strings. This observation does not establish the required result: An authorized release control is required before user-facing access changes.

D:

Tools must be discoverable with valid schemas before selection. This observation does not establish the required result: An authorized release control is required before user-facing access changes.

Question 105

An Azure-based AI team must satisfy this objective: A vision application receives a 429 throttling error during image analysis. Which design choice is best?

  1. Use region-aware captioning or a cropped image focused on that location
  2. Use the box coordinates to locate detected people in the source image
  3. Use evaluated confidence thresholds aligned to the business error cost
  4. Apply bounded retries and respect the service's throughput limit

Explanation

Correct Answer

D:

Too many concurrent requests can exceed resource quota. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Specific regions can carry details missed by a whole-image caption. That is not sufficient here: Too many concurrent requests can exceed resource quota.

B:

Localization metadata is tied to pixel regions, not just labels. That is not sufficient here: Too many concurrent requests can exceed resource quota.

C:

Confidence scores require calibration and contextual acceptance policy. That is not sufficient here: Too many concurrent requests can exceed resource quota.

Question 106

Troubleshooting report: The summary assigns all tasks to the first speaker. Which correction is MOST directly targeted to this failure?

  1. Replace the TTS stage with recognition/transcription
  2. Preserve speaker identities and confirm relevant transcript spans
  3. Combine vector similarity with structured catalog filtering
  4. Ground the summary in extracted transcript segments

Explanation

Correct Answer

B:

This correction directly addresses the reported failure. Attribution distinguishes who committed to each action. The targeted change is: Preserve speaker identities and confirm relevant transcript spans.

Incorrect Answers

A:

Speech recognition converts speech into words suitable for search. In the stated case, troubleshooting report: The summary assigns all tasks to the first speaker; “Replace the TTS stage with recognition/transcription” does not target that failure.

C:

Semantic vectors capture related concepts while filters enforce eligibility. In the stated case, troubleshooting report: The summary assigns all tasks to the first speaker; “Combine vector similarity with structured catalog filtering” does not target that failure.

D:

Meeting summaries require temporal and spoken source content. In the stated case, troubleshooting report: The summary assigns all tasks to the first speaker; “Ground the summary in extracted transcript segments” does not target that failure.

Question 107

A project needs the following AI behavior: A property assessor needs building defects marked directly on photographs. What should the designer implement?

  1. Identify a text-to-image generative workload, not an existing-image classifier
  2. Use named entity recognition with entity types
  3. Use image segmentation to map visual damage areas instead of assigning one global defect severity
  4. Use object detection when each product needs a location

Explanation

Correct Answer

C:

Pixel-level region masks localize building damage more precisely than an overall image class label. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Text-to-image generation creates new visual content from a description; classification only assigns labels to existing inputs. That is not sufficient here: Pixel-level region masks localize building damage more precisely than an overall image class label.

B:

NER locates and categorizes named entities in text. That is not sufficient here: Pixel-level region masks localize building damage more precisely than an overall image class label.

D:

Object detectors return classes and spatial boundaries. That is not sufficient here: Pixel-level region masks localize building damage more precisely than an overall image class label.

Question 108

An acceptance tester needs to validate: Use text-to-image generation and keep the input photo out of the edit request. Which observed result proves the intended behavior?

  1. The response distinguishes readable values from estimates
  2. A returned serial number is stored as OCR text not a generic image tag
  3. The answer reproduces the visible sign text accurately
  4. The result depicts the prompt without implying original photo preservation

Explanation

Correct Answer

D:

Generation from text is distinct from image editing. The expected acceptance evidence is: The result depicts the prompt without implying original photo preservation. This observation directly tests the stated behavior.

Incorrect Answers

A:

Visual reasoning is not proof of an exact value when axes are unclear. This observation does not establish the required result: The result depicts the prompt without implying original photo preservation.

B:

Tags describe concepts while OCR reports character content. This observation does not establish the required result: The result depicts the prompt without implying original photo preservation.

C:

Visual evidence is needed to read sign content. This observation does not establish the required result: The result depicts the prompt without implying original photo preservation.

Question 109

An Azure AI Fundamentals solution has this requirement: A client application sends prompts but never includes the user's actual question. Which implementation most directly addresses it?

  1. Check supported context and output limits for the chosen deployment
  2. Keep retrieved documents in lower-trust context and filter tool requests independently
  3. Place the current request in the supported user message field
  4. Configure the client with the actual resource endpoint and deployment identifier

Explanation

Correct Answer

C:

The inference service needs task content to generate a relevant answer. This implementation addresses the precise goal rather than an adjacent AI task.

Incorrect Answers

A:

Capabilities and limits vary by model deployment. That is not sufficient here: The inference service needs task content to generate a relevant answer.

B:

External content cannot replace privileged system behavior. That is not sufficient here: The inference service needs task content to generate a relevant answer.

D:

An inferencing request is routed to a concrete deployment configured in the resource. That is not sufficient here: The inference service needs task content to generate a relevant answer.

Question 110

Troubleshooting report: Every upload is forced through the invoice schema. Which correction is MOST directly targeted to this failure?

  1. Use entity recognition and verify sensitive-entity recall
  2. Call the authorized ticket tool and verify the receipt
  3. Introduce a document classifier and route by type
  4. Enforce backend identity and method scopes

Explanation

Correct Answer

C:

This correction directly addresses the reported failure. Classification selects the extraction process matching the document category. The targeted change is: Introduce a document classifier and route by type.

Incorrect Answers

A:

Entity extraction locates sensitive spans for privacy handling. In the stated case, troubleshooting report: Every upload is forced through the invoice schema; “Use entity recognition and verify sensitive-entity recall” does not target that failure.

B:

Agentic workloads combine reasoning and bounded external actions. In the stated case, troubleshooting report: Every upload is forced through the invoice schema; “Call the authorized ticket tool and verify the receipt” does not target that failure.

D:

Agentic workflows combine information access with controlled actions. In the stated case, troubleshooting report: Every upload is forced through the invoice schema; “Enforce backend identity and method scopes” does not target that failure.

Microsoft Azure AI Fundamentals (AI-901) Practice Tests