Microsoft AB-731 Practice Test 100
Question 990
At Willow Creek Finance, staff avoid AI because nobody explains where prompts are stored. Evidence: Conflicting messages describe data use and access controls. The business goal is to address data concerns through accurate policy communication. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- Chosen commercial model meets cost and performance requirements
- Employees accurately identify allowed inputs and escalation routes
- Employees can explain retained responsibility and tangible value
- An authorized person can explain and correct the outcome
Explanation
Correct Answer
B:
Conflicting messages describe data use and access controls. Employees accurately identify allowed inputs and escalation routes.
Incorrect Answers
A:
“Chosen commercial model meets cost and performance requirements” validates an outcome in a case where the evidence is: Demand is uncertain and production volume may be seasonal. This question instead requires an action that can address data concerns through accurate policy communication. Conflicting messages describe data use and access controls. Employees accurately identify allowed inputs and escalation routes.
C:
“Employees can explain retained responsibility and tangible value” validates an outcome in a case where the evidence is: Leadership communication focuses only on headcount cost savings. This question instead requires an action that can address data concerns through accurate policy communication. Conflicting messages describe data use and access controls. Employees accurately identify allowed inputs and escalation routes.
D:
“An authorized person can explain and correct the outcome” validates an outcome in a case where the evidence is: No human owner accepted responsibility for the decision. This question instead requires an action that can address data concerns through accurate policy communication. Conflicting messages describe data use and access controls. Employees accurately identify allowed inputs and escalation routes.
Question 991
At Bridgewater Logistics, a sales Copilot summary cites a confidential document shared too broadly. Evidence: Copilot follows existing file permissions, including mistaken grants. The business goal is to remediate repository permissions rather than hiding text in prompts. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- Colleagues can reuse it without seeing the original private information
- The model meets accepted quality on both old and new cohorts
- Spend decisions reflect measured benefit rather than raw adoption
- The previously unintended audience cannot retrieve the content
Explanation
Correct Answer
D:
Copilot follows existing file permissions, including mistaken grants. The previously unintended audience cannot retrieve the content.
Incorrect Answers
A:
“Colleagues can reuse it without seeing the original private information” validates an outcome in a case where the evidence is: A prompt may be reusable but embedded personal details are not. This question instead requires an action that can remediate repository permissions rather than hiding text in prompts. Copilot follows existing file permissions, including mistaken grants. The previously unintended audience cannot retrieve the content.
B:
“The model meets accepted quality on both old and new cohorts” validates an outcome in a case where the evidence is: New request types were absent from evaluation data. This question instead requires an action that can remediate repository permissions rather than hiding text in prompts. Copilot follows existing file permissions, including mistaken grants. The previously unintended audience cannot retrieve the content.
C:
“Spend decisions reflect measured benefit rather than raw adoption” validates an outcome in a case where the evidence is: License usage is not equivalent to measurable task improvement. This question instead requires an action that can remediate repository permissions rather than hiding text in prompts. Copilot follows existing file permissions, including mistaken grants. The previously unintended audience cannot retrieve the content.
Question 992
At Orion Distribution, a business unit purchases AI licenses but teams lack an accountable leader. Evidence: Employees receive mixed signals about priority and approved use. The business goal is to secure visible executive sponsorship with defined outcomes. The sponsor needs an evidence-based diagnosis rather than a guessed fix. Which initial investigation best identifies the controlling constraint before any solution is chosen?
- Identify sponsor, business goals and decision rights
- Identify baseline effort, incremental expenses and impact metrics
- Diagnose entitlement, sign-in, app version and tenant policy
- Identify executive sponsor, legal, security, HR and data stakeholders
Explanation
Correct Answer
A:
Employees receive mixed signals about priority and approved use. Identify sponsor, business goals and decision rights. Executive sponsorship supplies an accountable owner for priorities, resources and acceptable use. Without visibly endorsed outcomes, independent departments may adopt conflicting procedures or abandon the workflow.
Incorrect Answers
B:
“Identify baseline effort, incremental expenses and impact metrics” is a diagnostic step for the reported condition: Real costs include licenses, training, evaluation and human review. This question instead requires an action that can secure visible executive sponsorship with defined outcomes. Employees receive mixed signals about priority and approved use. Identify sponsor, business goals and decision rights. Executive sponsorship supplies an accountable owner for priorities, resources and acceptable use. Without visibly endorsed outcomes, independent departments may adopt conflicting procedures or abandon the workflow.
C:
“Diagnose entitlement, sign-in, app version and tenant policy” is a diagnostic step for the reported condition: Users do not know whether the issue is access, training or configuration. This question instead requires an action that can secure visible executive sponsorship with defined outcomes. Employees receive mixed signals about priority and approved use. Identify sponsor, business goals and decision rights. Executive sponsorship supplies an accountable owner for priorities, resources and acceptable use. Without visibly endorsed outcomes, independent departments may adopt conflicting procedures or abandon the workflow.
D:
“Identify executive sponsor, legal, security, HR and data stakeholders” is a diagnostic step for the reported condition: Different owners approve conflicting risk exceptions. This question instead requires an action that can secure visible executive sponsorship with defined outcomes. Employees receive mixed signals about priority and approved use. Identify sponsor, business goals and decision rights. Executive sponsorship supplies an accountable owner for priorities, resources and acceptable use. Without visibly endorsed outcomes, independent departments may adopt conflicting procedures or abandon the workflow.
Question 993
At Northlake Services, a proposed knowledge assistant sends an entire library to the model each time. Evidence: Large retrieved context increases token spending and response time. The business goal is to budget retrieval and generation costs together. The operating team needs an ongoing check after the initial rollout. Which post-rollout signal is most important for detecting regression or governance drift?
- Monitor retrieval cost per successful answer and quality regressions
- Monitor wide confidence intervals and missing difficult subgroups
- Monitor rubric drift and reviewer bias across updates
- Monitor copied confidential clauses and unauthorized prompts
Explanation
Correct Answer
A:
Large retrieved context increases token spending and response time. Monitor retrieval cost per successful answer and quality regressions.
Incorrect Answers
B:
“Monitor wide confidence intervals and missing difficult subgroups” monitors this signal: The sample cannot represent real volume and edge cases. This question instead requires an action that can budget retrieval and generation costs together. Large retrieved context increases token spending and response time. Monitor retrieval cost per successful answer and quality regressions.
C:
“Monitor rubric drift and reviewer bias across updates” monitors this signal: No documented rubric or blind comparison exists. This question instead requires an action that can budget retrieval and generation costs together. Large retrieved context increases token spending and response time. Monitor retrieval cost per successful answer and quality regressions.
D:
“Monitor copied confidential clauses and unauthorized prompts” monitors this signal: The user can access the approved summary but not full agreement text. This question instead requires an action that can budget retrieval and generation costs together. Large retrieved context increases token spending and response time. Monitor retrieval cost per successful answer and quality regressions.
Question 994
At Prairie Health, a ticket-processing assistant sometimes returns invalid JSON to a downstream parser. Evidence: An inconsistent structure causes automation failures. The business goal is to validate structured outputs rather than trusting formatting instructions. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- Inter-annotator agreement and minority-class performance improve
- All accepted payloads parse and missing required data is rejected
- Correct source fields appear in outputs after the data fix
- The system never overrides the deterministic reconciliation result
Explanation
Correct Answer
B:
An inconsistent structure causes automation failures. All accepted payloads parse and missing required data is rejected.
Incorrect Answers
A:
“Inter-annotator agreement and minority-class performance improve” validates an outcome in a case where the evidence is: Training labels are inconsistent and rare classes have few examples. This question instead requires an action that can validate structured outputs rather than trusting formatting instructions. An inconsistent structure causes automation failures. All accepted payloads parse and missing required data is rejected.
C:
“Correct source fields appear in outputs after the data fix” validates an outcome in a case where the evidence is: Retrieved field meanings changed even though model version is unchanged. This question instead requires an action that can validate structured outputs rather than trusting formatting instructions. An inconsistent structure causes automation failures. All accepted payloads parse and missing required data is rejected.
D:
“The system never overrides the deterministic reconciliation result” validates an outcome in a case where the evidence is: The reconciliation is deterministic arithmetic with known control rules. This question instead requires an action that can validate structured outputs rather than trusting formatting instructions. An inconsistent structure causes automation failures. All accepted payloads parse and missing required data is rejected.
Question 995
At Cascade Utilities, a communications group publishes an AI illustration as a photograph of an event. Evidence: The image may falsely imply a real situation occurred. The business goal is to require truthful labeling and suitability review for generated media. Only non-sensitive synthetic data and approved sandbox systems may be used. Which bounded trial with synthetic records best tests the proposed approach?
- Test mock publicity materials with one misleading caption
- Run a mock readiness exercise across three departments
- Test routine and exception cases in a mock workflow
- Run a fictional policy exception through the proposed council workflow
Explanation
Correct Answer
A:
The image may falsely imply a real situation occurred. Test mock publicity materials with one misleading caption.
Incorrect Answers
B:
“Run a mock readiness exercise across three departments” tests a proposed change for the condition: IT, HR, security and functional leaders run disconnected activities. This question instead requires an action that can require truthful labeling and suitability review for generated media. The image may falsely imply a real situation occurred. Test mock publicity materials with one misleading caption.
C:
“Test routine and exception cases in a mock workflow” tests a proposed change for the condition: Ambiguous policy exceptions can harm individuals. This question instead requires an action that can require truthful labeling and suitability review for generated media. The image may falsely imply a real situation occurred. Test mock publicity materials with one misleading caption.
D:
“Run a fictional policy exception through the proposed council workflow” tests a proposed change for the condition: Different owners approve conflicting risk exceptions. This question instead requires an action that can require truthful labeling and suitability review for generated media. The image may falsely imply a real situation occurred. Test mock publicity materials with one misleading caption.
Question 996
At Willow Creek Finance, a customer-support pilot reports five-star ratings from only satisfied users. Evidence: Feedback is self-selected and omits escalated failures. The business goal is to collect representative outcome evidence rather than vanity scores. The sponsor needs an evidence-based diagnosis rather than a guessed fix. Which initial investigation best identifies the controlling constraint before any solution is chosen?
- Identify which safe metadata can correlate a request with its output
- Record data residency, license, support and evaluation requirements
- Compare responder and non-responder populations and resolution rates
- Identify which steps are deterministic and which require interpretation
Explanation
Correct Answer
C:
Feedback is self-selected and omits escalated failures. Compare responder and non-responder populations and resolution rates. Voluntary feedback overrepresents people who chose to respond and can omit the most serious escalations. Link evaluation to representative task outcomes, failure logs and independent acceptance criteria.
Incorrect Answers
A:
“Identify which safe metadata can correlate a request with its output” is a diagnostic step for the reported condition: There is no trace from answer to model version and retrieved sources. This question instead requires an action that can collect representative outcome evidence rather than vanity scores. Feedback is self-selected and omits escalated failures. Compare responder and non-responder populations and resolution rates. Voluntary feedback overrepresents people who chose to respond and can omit the most serious escalations. Link evaluation to representative task outcomes, failure logs and independent acceptance criteria.
B:
“Record data residency, license, support and evaluation requirements” is a diagnostic step for the reported condition: Contractual hosting, customization and governance requirements differ. This question instead requires an action that can collect representative outcome evidence rather than vanity scores. Feedback is self-selected and omits escalated failures. Compare responder and non-responder populations and resolution rates. Voluntary feedback overrepresents people who chose to respond and can omit the most serious escalations. Link evaluation to representative task outcomes, failure logs and independent acceptance criteria.
D:
“Identify which steps are deterministic and which require interpretation” is a diagnostic step for the reported condition: The reconciliation is deterministic arithmetic with known control rules. This question instead requires an action that can collect representative outcome evidence rather than vanity scores. Feedback is self-selected and omits escalated failures. Compare responder and non-responder populations and resolution rates. Voluntary feedback overrepresents people who chose to respond and can omit the most serious escalations. Link evaluation to representative task outcomes, failure logs and independent acceptance criteria.
Question 997
At Cascade Utilities, a sales team needs instant short call tags and slower detailed briefs. Evidence: The two tasks have different turnaround and explanation requirements. The business goal is to use different quality-latency targets for distinct AI workflows. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- No unauthorized action follows the retrieved directive
- Predicted categories agree with approved labels and uncertain items escalate
- Both task classes meet their own latency and quality gates
- Inter-rater agreement and quality meet predefined thresholds
Explanation
Correct Answer
C:
The two tasks have different turnaround and explanation requirements. Both task classes meet their own latency and quality gates. A quick status response and a defensible compliance analysis need different latency, evidence and review budgets. Set separate acceptance targets rather than optimize all requests for one average speed.
Incorrect Answers
A:
“No unauthorized action follows the retrieved directive” validates an outcome in a case where the evidence is: The embedded directive came from lower-trust document content. This question instead requires an action that can use different quality-latency targets for distinct AI workflows. The two tasks have different turnaround and explanation requirements. Both task classes meet their own latency and quality gates. A quick status response and a defensible compliance analysis need different latency, evidence and review budgets. Set separate acceptance targets rather than optimize all requests for one average speed.
B:
“Predicted categories agree with approved labels and uncertain items escalate” validates an outcome in a case where the evidence is: The desired answer is a stable label rather than a free-form narrative. This question instead requires an action that can use different quality-latency targets for distinct AI workflows. The two tasks have different turnaround and explanation requirements. Both task classes meet their own latency and quality gates. A quick status response and a defensible compliance analysis need different latency, evidence and review budgets. Set separate acceptance targets rather than optimize all requests for one average speed.
D:
“Inter-rater agreement and quality meet predefined thresholds” validates an outcome in a case where the evidence is: No documented rubric or blind comparison exists. This question instead requires an action that can use different quality-latency targets for distinct AI workflows. The two tasks have different turnaround and explanation requirements. Both task classes meet their own latency and quality gates. A quick status response and a defensible compliance analysis need different latency, evidence and review budgets. Set separate acceptance targets rather than optimize all requests for one average speed.
Question 998
At Solstice Retail, a service desk wants an agent grounded in approved return-policy articles. Evidence: The questions recur and answers must reflect maintained policy knowledge. The business goal is to use Copilot Studio for a governed agent when a standard chat is insufficient. The operating team needs an ongoing check after the initial rollout. Which post-rollout signal is most important for detecting regression or governance drift?
- Monitor duplicate agents and maintenance burden
- Monitor exact-id misses and semantic false positives
- Monitor stale knowledge sources and unsupported decisions
- Monitor broken links and unstated workbook revisions
Explanation
Correct Answer
C:
The questions recur and answers must reflect maintained policy knowledge. Monitor stale knowledge sources and unsupported decisions.
Incorrect Answers
A:
“Monitor duplicate agents and maintenance burden” monitors this signal: A duplicate custom project adds governance and maintenance cost. This question instead requires an action that can use Copilot Studio for a governed agent when a standard chat is insufficient. The questions recur and answers must reflect maintained policy knowledge. Monitor stale knowledge sources and unsupported decisions.
B:
“Monitor exact-id misses and semantic false positives” monitors this signal: Vector similarity alone is not enough to preserve identifiers. This question instead requires an action that can use Copilot Studio for a governed agent when a standard chat is insufficient. The questions recur and answers must reflect maintained policy knowledge. Monitor stale knowledge sources and unsupported decisions.
D:
“Monitor broken links and unstated workbook revisions” monitors this signal: Some underlying work records were revised after the first draft. This question instead requires an action that can use Copilot Studio for a governed agent when a standard chat is insufficient. The questions recur and answers must reflect maintained policy knowledge. Monitor stale knowledge sources and unsupported decisions.
Question 999
At Lakeshore Commerce, a retrieved file tells the assistant to ignore its corporate instructions. Evidence: The embedded directive came from lower-trust document content. The business goal is to separate trusted system intent from retrieved text. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- No unauthorized action follows the retrieved directive
- The pilot meets predeclared performance and risk gates before expansion
- Extracted amounts and row associations match source documents
- All accepted payloads parse and missing required data is rejected
Explanation
Correct Answer
A:
The embedded directive came from lower-trust document content. No unauthorized action follows the retrieved directive.
Incorrect Answers
B:
“The pilot meets predeclared performance and risk gates before expansion” validates an outcome in a case where the evidence is: High deployment effort conceals whether one use case can succeed. This question instead requires an action that can separate trusted system intent from retrieved text. The embedded directive came from lower-trust document content. No unauthorized action follows the retrieved directive.
C:
“Extracted amounts and row associations match source documents” validates an outcome in a case where the evidence is: Corrupt source extraction can make even a grounded answer wrong. This question instead requires an action that can separate trusted system intent from retrieved text. The embedded directive came from lower-trust document content. No unauthorized action follows the retrieved directive.
D:
“All accepted payloads parse and missing required data is rejected” validates an outcome in a case where the evidence is: An inconsistent structure causes automation failures. This question instead requires an action that can separate trusted system intent from retrieved text. The embedded directive came from lower-trust document content. No unauthorized action follows the retrieved directive.
Question 1000
At Marigold Education, a maintenance operator has images and brief text notes of equipment defects. Evidence: Text descriptions alone omit visual indicators needed for triage. The business goal is to evaluate vision-capable AI when image evidence materially matters. The sponsor needs an evidence-based diagnosis rather than a guessed fix. Which initial investigation best identifies the controlling constraint before any solution is chosen?
- Check requester identity, entitlement and purpose for exact text
- Compare responder and non-responder populations and resolution rates
- Confirm whether failures result from stale facts rather than response style
- Check image quality, permission and what decision needs visual cues
Explanation
Correct Answer
D:
Text descriptions alone omit visual indicators needed for triage. Check image quality, permission and what decision needs visual cues. A textual description cannot confirm damage, layout or other cues visible only in an image. Evaluate an approved vision-capable system on labeled examples and keep ambiguous findings for human review.
Incorrect Answers
A:
“Check requester identity, entitlement and purpose for exact text” is a diagnostic step for the reported condition: The user can access the approved summary but not full agreement text. This question instead requires an action that can evaluate vision-capable AI when image evidence materially matters. Text descriptions alone omit visual indicators needed for triage. Check image quality, permission and what decision needs visual cues. A textual description cannot confirm damage, layout or other cues visible only in an image. Evaluate an approved vision-capable system on labeled examples and keep ambiguous findings for human review.
B:
“Compare responder and non-responder populations and resolution rates” is a diagnostic step for the reported condition: Feedback is self-selected and omits escalated failures. This question instead requires an action that can evaluate vision-capable AI when image evidence materially matters. Text descriptions alone omit visual indicators needed for triage. Check image quality, permission and what decision needs visual cues. A textual description cannot confirm damage, layout or other cues visible only in an image. Evaluate an approved vision-capable system on labeled examples and keep ambiguous findings for human review.
C:
“Confirm whether failures result from stale facts rather than response style” is a diagnostic step for the reported condition: A static model snapshot cannot reliably know the latest approved document. This question instead requires an action that can evaluate vision-capable AI when image evidence materially matters. Text descriptions alone omit visual indicators needed for triage. Check image quality, permission and what decision needs visual cues. A textual description cannot confirm damage, layout or other cues visible only in an image. Evaluate an approved vision-capable system on labeled examples and keep ambiguous findings for human review.