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 57

Question 560

Troubleshooting report: The model invents a required parameter name not in the contract. Which correction is MOST directly targeted to this failure?

  1. Constrain the call to the documented JSON input schema
  2. Partition conversation context by identity and expiration
  3. Exercise negative and expected tool paths before release
  4. Require real tool calls and distinguish pending from completed

Explanation

Correct Answer

A:

This correction directly addresses the reported failure. Invalid tool parameters break calls even if the intended operation is correct. The targeted change is: Constrain the call to the documented JSON input schema.

Incorrect Answers

B:

Conversation continuity should preserve identifiers without crossing user sessions. In the stated case, troubleshooting report: The model invents a required parameter name not in the contract; “Partition conversation context by identity and expiration” does not target that failure.

C:

An answer that sounds plausible may conceal failed tool calls. In the stated case, troubleshooting report: The model invents a required parameter name not in the contract; “Exercise negative and expected tool paths before release” does not target that failure.

D:

Generated text does not prove external execution. In the stated case, troubleshooting report: The model invents a required parameter name not in the contract; “Require real tool calls and distinguish pending from completed” does not target that failure.

Question 561

The approved design action is: Put persistent task boundaries in a supported high-priority system instruction. Which test result would BEST establish that it worked?

  1. An out-of-scope request is handled under the approved restrictions
  2. A request using the deployment name returns a response from that resource
  3. Injection attempts fail to access protected secrets or tools
  4. The second tenant's test passes with its own authorized identity

Explanation

Correct Answer

A:

High-priority instructions should describe approved behavior and escalation rules. The expected acceptance evidence is: An out-of-scope request is handled under the approved restrictions. This observation directly tests the stated behavior.

Incorrect Answers

B:

An inferencing request is routed to a concrete deployment configured in the resource. This observation does not establish the required result: An out-of-scope request is handled under the approved restrictions.

C:

A user cannot use text to elevate permissions beyond the application policy. This observation does not establish the required result: An out-of-scope request is handled under the approved restrictions.

D:

Copying code does not grant access to another tenant's resources. This observation does not establish the required result: An out-of-scope request is handled under the approved restrictions.

Question 562

Troubleshooting report: Safety notifications arrive in an unmonitored shared mailbox. Which correction is MOST directly targeted to this failure?

  1. Gate the payment tool with verified role and amount checks
  2. Apply data minimization, retention and permission controls
  3. Create an on-call ownership and escalation schedule
  4. Mask or exclude unauthorized identity details before publishing

Explanation

Correct Answer

C:

This correction directly addresses the reported failure. An AI model cannot be its own accountable incident responder. The targeted change is: Create an on-call ownership and escalation schedule.

Incorrect Answers

A:

A prompt-level request cannot override a transaction policy. In the stated case, troubleshooting report: Safety notifications arrive in an unmonitored shared mailbox; “Gate the payment tool with verified role and amount checks” does not target that failure.

B:

Privacy requires meaningful notice and bounded handling of personal data. In the stated case, troubleshooting report: Safety notifications arrive in an unmonitored shared mailbox; “Apply data minimization, retention and permission controls” does not target that failure.

D:

Privacy applies to image content as well as text transcripts. In the stated case, troubleshooting report: Safety notifications arrive in an unmonitored shared mailbox; “Mask or exclude unauthorized identity details before publishing” does not target that failure.

Question 563

Troubleshooting report: An expired secret is embedded in the application configuration. Which correction is MOST directly targeted to this failure?

  1. Use the supported image-content input and multimodal deployment
  2. Use managed identity or rotated approved credentials and correct role assignments
  3. Reinforce role boundaries and enforce secrets outside prompts
  4. Rotate the key and load credentials through a governed secret mechanism

Explanation

Correct Answer

B:

This correction directly addresses the reported failure. Authentication proves caller identity and authorization grants resource access. The targeted change is: Use managed identity or rotated approved credentials and correct role assignments.

Incorrect Answers

A:

The model must receive the actual visual content to reason about it. In the stated case, troubleshooting report: An expired secret is embedded in the application configuration; “Use the supported image-content input and multimodal deployment” does not target that failure.

C:

Developer and system constraints govern allowed operations over user requests. In the stated case, troubleshooting report: An expired secret is embedded in the application configuration; “Reinforce role boundaries and enforce secrets outside prompts” does not target that failure.

D:

Credentials should not appear in repositories, prompt logs or client binaries. In the stated case, troubleshooting report: An expired secret is embedded in the application configuration; “Rotate the key and load credentials through a governed secret mechanism” does not target that failure.

Question 564

Troubleshooting report: The application uses a key from a different resource. Which correction is MOST directly targeted to this failure?

  1. Choose the correct TTS voice from the supported catalog
  2. Add realistic noise tests and a clarification fallback
  3. Supply an authorized identity and matching resource configuration
  4. Invoke a text-to-speech voice and compatible audio format

Explanation

Correct Answer

C:

This correction directly addresses the reported failure. Audio capture success does not establish remote service authentication. The targeted change is: Supply an authorized identity and matching resource configuration.

Incorrect Answers

A:

Voice timbre is a synthesis configuration, not speech recognition output. In the stated case, troubleshooting report: The application uses a key from a different resource; “Choose the correct TTS voice from the supported catalog” does not target that failure.

B:

Signal quality directly affects recognizer error rates. In the stated case, troubleshooting report: The application uses a key from a different resource; “Add realistic noise tests and a clarification fallback” does not target that failure.

D:

Speech synthesis generates audio waveforms from supplied text. In the stated case, troubleshooting report: The application uses a key from a different resource; “Invoke a text-to-speech voice and compatible audio format” does not target that failure.

Question 565

The approved design action is: Use text-to-speech with a supported voice and accessible controls. Which test result would BEST establish that it worked?

  1. Follow-up questions resolve consistently to the same approved case documents
  2. Held-out events produce measured recall and precision for failures
  3. The spoken message matches the approved written text
  4. The minutes correctly associate tasks with speakers and source timestamps

Explanation

Correct Answer

C:

Speech synthesis converts supplied text into audible output. The expected acceptance evidence is: The spoken message matches the approved written text. This observation directly tests the stated behavior.

Incorrect Answers

A:

The assistant should incorporate context while remaining connected to reliable sources. This observation does not establish the required result: The spoken message matches the approved written text.

B:

A failure probability or class is a discrete decision under a defined horizon. This observation does not establish the required result: The spoken message matches the approved written text.

D:

Attribution distinguishes who committed to each action. This observation does not establish the required result: The spoken message matches the approved written text.

Question 566

An Azure AI solution uses the following approved control: Set authorized knowledge sources and a no-source fallback. What observable result should pass QA?

  1. A trace shows which tool result supported each claimed business action
  2. Questions without a matching manual passage yield a qualified response
  3. A list operation succeeds and a mutation attempt is denied
  4. An auditor can inspect the saved response alongside the exact prompt and model configuration used

Explanation

Correct Answer

B:

Retrieval access and abstention avoid inventing unsupported documentation. The expected acceptance evidence is: Questions without a matching manual passage yield a qualified response. This observation directly tests the stated behavior.

Incorrect Answers

A:

Operational observability aids debugging without exposing hidden sensitive prompt data. This observation does not establish the required result: Questions without a matching manual passage yield a qualified response.

C:

Least-privilege tool methods reduce accidental external actions. This observation does not establish the required result: Questions without a matching manual passage yield a qualified response.

D:

Recorded settings support investigation, but generative models can still vary on an apparently identical rerun. This observation does not establish the required result: Questions without a matching manual passage yield a qualified response.

Question 567

A solution team has selected this control: Represent text with embeddings and compare vectors using an appropriate similarity measure. Which observation is the most direct acceptance evidence?

  1. Semantically related sentences have nearer vectors in an evaluated sample
  2. The resulting cohorts are evaluated for cohesion and business usefulness
  3. Queries and indexed vectors carry matching model-version metadata
  4. The smaller candidate meets the acceptance floor with lower cost where appropriate

Explanation

Correct Answer

A:

Embeddings encode learned semantic relationships rather than literal string identity. The expected acceptance evidence is: Semantically related sentences have nearer vectors in an evaluated sample. This observation directly tests the stated behavior.

Incorrect Answers

B:

Unsupervised learning finds structure without supplied target labels. This observation does not establish the required result: Semantically related sentences have nearer vectors in an evaluated sample.

C:

Incompatible vector spaces can make nearest-neighbor rankings meaningless. This observation does not establish the required result: Semantically related sentences have nearer vectors in an evaluated sample.

D:

Model size alone does not establish suitability for a particular task. This observation does not establish the required result: Semantically related sentences have nearer vectors in an evaluated sample.

Question 568

A solution team has selected this control: Choose detection only if spatial object locations are required. Which observation is the most direct acceptance evidence?

  1. A multi-object test returns boxes when location is required
  2. The test workload recovers without unbounded retry bursts
  3. Prominent objects receive separate region captions and coordinates
  4. The output is a validated outdoor probability or category

Explanation

Correct Answer

A:

Classification predicts image-level categories while detection provides positions. The expected acceptance evidence is: A multi-object test returns boxes when location is required. This observation directly tests the stated behavior.

Incorrect Answers

B:

Too many concurrent requests can exceed resource quota. This observation does not establish the required result: A multi-object test returns boxes when location is required.

C:

Dense captions describe localized regions beyond a single global caption. This observation does not establish the required result: A multi-object test returns boxes when location is required.

D:

Image-level classification is distinct from object localization. This observation does not establish the required result: A multi-object test returns boxes when location is required.

Question 569

Troubleshooting report: The app changes a language-detection setting expecting a different voice. Which correction is MOST directly targeted to this failure?

  1. Ground the summary and penalize unsupported additions
  2. Gate side effects on final recognition and explicit confirmation
  3. Choose the correct TTS voice from the supported catalog
  4. Enable supported diarization and validate turn assignments

Explanation

Correct Answer

C:

This correction directly addresses the reported failure. Voice timbre is a synthesis configuration, not speech recognition output. The targeted change is: Choose the correct TTS voice from the supported catalog.

Incorrect Answers

A:

Summarization compresses written information rather than changing its facts. In the stated case, troubleshooting report: The app changes a language-detection setting expecting a different voice; “Ground the summary and penalize unsupported additions” does not target that failure.

B:

Interim hypotheses can change as more audio arrives. In the stated case, troubleshooting report: The app changes a language-detection setting expecting a different voice; “Gate side effects on final recognition and explicit confirmation” does not target that failure.

D:

Speaker diarization separates turns and identities at a conversation level. In the stated case, troubleshooting report: The app changes a language-detection setting expecting a different voice; “Enable supported diarization and validate turn assignments” does not target that failure.

Question 570

An Azure AI solution uses the following approved control: Use validated data retrieval and source-backed numerical responses. What observable result should pass QA?

  1. A rejected test case can be reviewed with its supporting evidence
  2. Eligible memory entries disappear from active stores after the verified request
  3. A first-use test displays clear AI identification and human contact option
  4. The returned metric reconciles to the approved source table

Explanation

Correct Answer

D:

A numeric-looking answer is not proof that a measure was computed. The expected acceptance evidence is: The returned metric reconciles to the approved source table. This observation directly tests the stated behavior.

Incorrect Answers

A:

Accountability includes traceable responsibility and appropriate contestability. This observation does not establish the required result: The returned metric reconciles to the approved source table.

B:

Accountability requires evidence of actual data lifecycle actions. This observation does not establish the required result: The returned metric reconciles to the approved source table.

C:

Transparency helps users assess the scope and limitations of the assistance. This observation does not establish the required result: The returned metric reconciles to the approved source table.

Microsoft Azure AI Fundamentals (AI-901) Practice Tests