Microsoft AB-731 Practice Test 50
Question 490
At Ashwood Public Services, an AI council reports success only as the number of policies issued. Evidence: No evidence links governance to fewer incidents or better outcomes. The business goal is to measure governance effectiveness with outcome indicators. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- Leaders prioritize actual risk reduction rather than paperwork volume
- Unsafe situations trigger safe handoff rather than fabricated instructions
- The rollout gate blocks access risks before launch
- Decisions show owners, recorded rationale and appeal routes
Explanation
Correct Answer
A:
No evidence links governance to fewer incidents or better outcomes. Leaders prioritize actual risk reduction rather than paperwork volume.
Incorrect Answers
B:
“Unsafe situations trigger safe handoff rather than fabricated instructions” validates an outcome in a case where the evidence is: Incorrect advice could cause physical harm. This question instead requires an action that can measure governance effectiveness with outcome indicators. No evidence links governance to fewer incidents or better outcomes. Leaders prioritize actual risk reduction rather than paperwork volume.
C:
“The rollout gate blocks access risks before launch” validates an outcome in a case where the evidence is: Enterprise repositories contain broad anonymous sharing links. This question instead requires an action that can measure governance effectiveness with outcome indicators. No evidence links governance to fewer incidents or better outcomes. Leaders prioritize actual risk reduction rather than paperwork volume.
D:
“Decisions show owners, recorded rationale and appeal routes” validates an outcome in a case where the evidence is: Different owners approve conflicting risk exceptions. This question instead requires an action that can measure governance effectiveness with outcome indicators. No evidence links governance to fewer incidents or better outcomes. Leaders prioritize actual risk reduction rather than paperwork volume.
Question 491
At Coral Bay Systems, a claims classification model performs badly on unusual claim types. Evidence: Training labels are inconsistent and rare classes have few examples. The business goal is to improve labeling quality and representation before model tuning. 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 unnecessary health fields in each request payload
- Separate short routing tasks from analysis-heavy summaries
- Sample false positives and audit label definitions by category
- Review endpoint dependencies, timeouts and critical workflow priority
Explanation
Correct Answer
C:
Training labels are inconsistent and rare classes have few examples. Sample false positives and audit label definitions by category. Inconsistent labels teach the model contradictory targets, while missing rare classes distort accuracy. Resolve annotation rules and sample coverage before changing architecture or tuning hyperparameters.
Incorrect Answers
A:
“Identify unnecessary health fields in each request payload” is a diagnostic step for the reported condition: Only scheduling constraints and contact details are needed. This question instead requires an action that can improve labeling quality and representation before model tuning. Training labels are inconsistent and rare classes have few examples. Sample false positives and audit label definitions by category. Inconsistent labels teach the model contradictory targets, while missing rare classes distort accuracy. Resolve annotation rules and sample coverage before changing architecture or tuning hyperparameters.
B:
“Separate short routing tasks from analysis-heavy summaries” is a diagnostic step for the reported condition: The two tasks have different turnaround and explanation requirements. This question instead requires an action that can improve labeling quality and representation before model tuning. Training labels are inconsistent and rare classes have few examples. Sample false positives and audit label definitions by category. Inconsistent labels teach the model contradictory targets, while missing rare classes distort accuracy. Resolve annotation rules and sample coverage before changing architecture or tuning hyperparameters.
D:
“Review endpoint dependencies, timeouts and critical workflow priority” is a diagnostic step for the reported condition: A single dependent service outage stops all customer help. This question instead requires an action that can improve labeling quality and representation before model tuning. Training labels are inconsistent and rare classes have few examples. Sample false positives and audit label definitions by category. Inconsistent labels teach the model contradictory targets, while missing rare classes distort accuracy. Resolve annotation rules and sample coverage before changing architecture or tuning hyperparameters.
Question 492
At Sapphire Mobility, a bank purchases an AI service without reviewing vendor processing terms. Evidence: Contract data and prompts may be handled under unclear commitments. The business goal is to review third-party AI risks before procurement approval. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- The application declines unauthorized requests and logs attempts
- Observed differences align with legitimate criteria, not proxy bias
- The selected supplier meets mandatory governance controls
- Learners demonstrate safe task proficiency rather than attendance alone
Explanation
Correct Answer
C:
Contract data and prompts may be handled under unclear commitments. The selected supplier meets mandatory governance controls.
Incorrect Answers
A:
“The application declines unauthorized requests and logs attempts” validates an outcome in a case where the evidence is: Malicious or unexpected inputs could bypass intended restrictions. This question instead requires an action that can review third-party AI risks before procurement approval. Contract data and prompts may be handled under unclear commitments. The selected supplier meets mandatory governance controls.
B:
“Observed differences align with legitimate criteria, not proxy bias” validates an outcome in a case where the evidence is: Proxies may create discriminatory patterns even without explicit protected fields. This question instead requires an action that can review third-party AI risks before procurement approval. Contract data and prompts may be handled under unclear commitments. The selected supplier meets mandatory governance controls.
D:
“Learners demonstrate safe task proficiency rather than attendance alone” is a design or implementation response to the condition: Nontechnical teams need practical role-specific workflows. This question instead requires an action that can review third-party AI risks before procurement approval. Contract data and prompts may be handled under unclear commitments. The selected supplier meets mandatory governance controls.
Question 493
At Cedar Analytics, a consumer company expects tenfold seasonal chat volume. Evidence: The current manual process cannot expand rapidly without hiring. The business goal is to evaluate whether a bounded AI workflow can scale safely. 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 tool authorization attempts and scope escalation
- Monitor transcription errors, consent settings and missing audio segments
- Monitor unauthorized approvals and escalation backlog
- Watch capacity saturation, failed handoffs and unit cost under peaks
Explanation
Correct Answer
D:
The current manual process cannot expand rapidly without hiring. Watch capacity saturation, failed handoffs and unit cost under peaks.
Incorrect Answers
A:
“Monitor tool authorization attempts and scope escalation” monitors this signal: Retrieving an account statement does not authorize transactions. This question instead requires an action that can evaluate whether a bounded AI workflow can scale safely. The current manual process cannot expand rapidly without hiring. Watch capacity saturation, failed handoffs and unit cost under peaks.
B:
“Monitor transcription errors, consent settings and missing audio segments” monitors this signal: Poor transcription can propagate incorrect speaker attribution into summaries. This question instead requires an action that can evaluate whether a bounded AI workflow can scale safely. The current manual process cannot expand rapidly without hiring. Watch capacity saturation, failed handoffs and unit cost under peaks.
C:
“Monitor unauthorized approvals and escalation backlog” monitors this signal: Decisions have legal consequences and policy ambiguity. This question instead requires an action that can evaluate whether a bounded AI workflow can scale safely. The current manual process cannot expand rapidly without hiring. Watch capacity saturation, failed handoffs and unit cost under peaks.
Question 494
At Riverbend Publishing, a compliance team needs both matching clauses and exact regulation numbers. Evidence: Vector similarity alone is not enough to preserve identifiers. The business goal is to choose retrieval matching both meaning and exact references. The proposed solution must respect documented authority and organizational controls. Which approach most directly meets the stated business objective within the approved authority?
- Add groundedness criteria to the model evaluation and release gate
- Use index fields supporting lexical and vector search
- Draft in Outlook with verified language and require manual sending
- Block release until model and application mitigations pass tests
Explanation
Correct Answer
B:
Vector similarity alone is not enough to preserve identifiers. Use index fields supporting lexical and vector search.
Incorrect Answers
A:
“Add groundedness criteria to the model evaluation and release gate” is a design or implementation response to the condition: Citations exist but conclusions are not entailed by the passages. This question instead requires an action that can choose retrieval matching both meaning and exact references. Vector similarity alone is not enough to preserve identifiers. Use index fields supporting lexical and vector search.
C:
“Draft in Outlook with verified language and require manual sending” tests a proposed change for the condition: A proposed discount is mentioned but not yet approved. This question instead requires an action that can choose retrieval matching both meaning and exact references. Vector similarity alone is not enough to preserve identifiers. Use index fields supporting lexical and vector search.
D:
“Block release until model and application mitigations pass tests” is a design or implementation response to the condition: Performance alone does not demonstrate responsible deployment. This question instead requires an action that can choose retrieval matching both meaning and exact references. Vector similarity alone is not enough to preserve identifiers. Use index fields supporting lexical and vector search.
Question 495
At Juniper Operations, 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 proposed solution must respect documented authority and organizational controls. Which approach most directly meets the stated business objective within the approved authority?
- Use a sanctioned connection or authorized export rather than inventing records
- Review Copilot recap and flag unconfirmed commitments
- Have the document owner correct access and revalidate visibility
- Use synthetic or authorized redacted images with approved retention
Explanation
Correct Answer
C:
Copilot follows existing file permissions, including mistaken grants. Have the document owner correct access and revalidate visibility.
Incorrect Answers
A:
“Use a sanctioned connection or authorized export rather than inventing records” tests a proposed change for the condition: The data is not necessarily available to Microsoft 365 Copilot without configuration. 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. Have the document owner correct access and revalidate visibility.
B:
“Review Copilot recap and flag unconfirmed commitments” is a design or implementation response to the condition: Transcript includes suggestions that were not formally accepted. 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. Have the document owner correct access and revalidate visibility.
D:
“Use synthetic or authorized redacted images with approved retention” tests a proposed change for the condition: A less sensitive evaluation could answer the technical question. 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. Have the document owner correct access and revalidate visibility.
Question 496
At Lakeshore Commerce, 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 proposed solution must respect documented authority and organizational controls. Which approach most directly meets the stated business objective within the approved authority?
- Assign evaluated response modes and service-level objectives by task type
- Store versioned prompts with owner, regression tests and rollback
- Combine sampled quality audits with user feedback and escalation metrics
- Index embeddings for matching and use a generator only after retrieval
Explanation
Correct Answer
C:
Feedback is self-selected and omits escalated failures. Combine sampled quality audits with user feedback and escalation metrics. 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:
“Assign evaluated response modes and service-level objectives by task type” is a design or implementation response to the condition: The two tasks have different turnaround and explanation requirements. This question instead requires an action that can collect representative outcome evidence rather than vanity scores. Feedback is self-selected and omits escalated failures. Combine sampled quality audits with user feedback and escalation metrics. 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:
“Store versioned prompts with owner, regression tests and rollback” is a design or implementation response to the condition: Prompt templates have no version owner or change record. This question instead requires an action that can collect representative outcome evidence rather than vanity scores. Feedback is self-selected and omits escalated failures. Combine sampled quality audits with user feedback and escalation metrics. 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:
“Index embeddings for matching and use a generator only after retrieval” is a design or implementation response to the condition: Similarity matching is distinct from composing a final recommendation. This question instead requires an action that can collect representative outcome evidence rather than vanity scores. Feedback is self-selected and omits escalated failures. Combine sampled quality audits with user feedback and escalation metrics. 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 497
At Clearwater Partners, a loan assessment model omits race but uses highly correlated location variables. Evidence: Proxies may create discriminatory patterns even without explicit protected fields. The business goal is to test fairness effects beyond removal of direct identifiers. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- Summary captures decisions without exposing irrelevant private information
- Decisions show owners, recorded rationale and appeal routes
- The rollout gate blocks access risks before launch
- Observed differences align with legitimate criteria, not proxy bias
Explanation
Correct Answer
D:
Proxies may create discriminatory patterns even without explicit protected fields. Observed differences align with legitimate criteria, not proxy bias. Variables correlated with protected characteristics can preserve unequal outcomes after explicit identifiers are removed. Cohort-level testing and investigation of the influencing features are needed to identify proxy effects.
Incorrect Answers
A:
“Summary captures decisions without exposing irrelevant private information” validates an outcome in a case where the evidence is: The project only needs agreed project decisions. This question instead requires an action that can test fairness effects beyond removal of direct identifiers. Proxies may create discriminatory patterns even without explicit protected fields. Observed differences align with legitimate criteria, not proxy bias. Variables correlated with protected characteristics can preserve unequal outcomes after explicit identifiers are removed. Cohort-level testing and investigation of the influencing features are needed to identify proxy effects.
B:
“Decisions show owners, recorded rationale and appeal routes” validates an outcome in a case where the evidence is: Different owners approve conflicting risk exceptions. This question instead requires an action that can test fairness effects beyond removal of direct identifiers. Proxies may create discriminatory patterns even without explicit protected fields. Observed differences align with legitimate criteria, not proxy bias. Variables correlated with protected characteristics can preserve unequal outcomes after explicit identifiers are removed. Cohort-level testing and investigation of the influencing features are needed to identify proxy effects.
C:
“The rollout gate blocks access risks before launch” validates an outcome in a case where the evidence is: Enterprise repositories contain broad anonymous sharing links. This question instead requires an action that can test fairness effects beyond removal of direct identifiers. Proxies may create discriminatory patterns even without explicit protected fields. Observed differences align with legitimate criteria, not proxy bias. Variables correlated with protected characteristics can preserve unequal outcomes after explicit identifiers are removed. Cohort-level testing and investigation of the influencing features are needed to identify proxy effects.
Question 498
At Orion Distribution, an enterprise RAG index contains a malicious imported document. Evidence: The document includes text telling the assistant to export hidden records. The business goal is to treat retrieved content as untrusted and isolate tool permissions. 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 model catalog, evaluation, deployment and security requirements
- Inspect data structure, numeric units and expected computations
- Record token volumes, retries and actual chosen billing meters
- Find lower-trust instructions in retrieved passages
Explanation
Correct Answer
D:
The document includes text telling the assistant to export hidden records. Find lower-trust instructions in retrieved passages. Instructions inside retrieved documents are untrusted data, not authorization to call tools or expose records. Enforce tool permissions outside the model and treat attempted instruction redirection as hostile content.
Incorrect Answers
A:
“Identify model catalog, evaluation, deployment and security requirements” is a diagnostic step for the reported condition: The team must compare models, tools and safety controls as one lifecycle. This question instead requires an action that can treat retrieved content as untrusted and isolate tool permissions. The document includes text telling the assistant to export hidden records. Find lower-trust instructions in retrieved passages. Instructions inside retrieved documents are untrusted data, not authorization to call tools or expose records. Enforce tool permissions outside the model and treat attempted instruction redirection as hostile content.
B:
“Inspect data structure, numeric units and expected computations” tests a proposed change for the condition: The task requires calculations and statistical checks beyond simple prose. This question instead requires an action that can treat retrieved content as untrusted and isolate tool permissions. The document includes text telling the assistant to export hidden records. Find lower-trust instructions in retrieved passages. Instructions inside retrieved documents are untrusted data, not authorization to call tools or expose records. Enforce tool permissions outside the model and treat attempted instruction redirection as hostile content.
C:
“Record token volumes, retries and actual chosen billing meters” is a diagnostic step for the reported condition: Input and output tokens and deployment type all affect billing. This question instead requires an action that can treat retrieved content as untrusted and isolate tool permissions. The document includes text telling the assistant to export hidden records. Find lower-trust instructions in retrieved passages. Instructions inside retrieved documents are untrusted data, not authorization to call tools or expose records. Enforce tool permissions outside the model and treat attempted instruction redirection as hostile content.
Question 499
At BrightPath Media, an international company wants a single AI adoption policy for all jurisdictions. Evidence: Availability, privacy law and employee consultation rules differ by region. The business goal is to adapt rollout controls to local obligations without lowering core safety. 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 repeated blockers and outdated troubleshooting articles
- Monitor disparate impact and classification drift
- Monitor support load and quality loss with expanded usage
- Monitor regulatory changes and unapproved cross-border transfers
Explanation
Correct Answer
D:
Availability, privacy law and employee consultation rules differ by region. Monitor regulatory changes and unapproved cross-border transfers. A feature or data practice acceptable in one jurisdiction may require different notice, consultation or deployment controls elsewhere. Review local obligations while preserving consistent minimum safeguards.
Incorrect Answers
A:
“Monitor repeated blockers and outdated troubleshooting articles” monitors this signal: Users do not know whether the issue is access, training or configuration. This question instead requires an action that can adapt rollout controls to local obligations without lowering core safety. Availability, privacy law and employee consultation rules differ by region. Monitor regulatory changes and unapproved cross-border transfers. A feature or data practice acceptable in one jurisdiction may require different notice, consultation or deployment controls elsewhere. Review local obligations while preserving consistent minimum safeguards.
B:
“Monitor disparate impact and classification drift” monitors this signal: Performance differs sharply across legally relevant applicant groups. This question instead requires an action that can adapt rollout controls to local obligations without lowering core safety. Availability, privacy law and employee consultation rules differ by region. Monitor regulatory changes and unapproved cross-border transfers. A feature or data practice acceptable in one jurisdiction may require different notice, consultation or deployment controls elsewhere. Review local obligations while preserving consistent minimum safeguards.
C:
“Monitor support load and quality loss with expanded usage” monitors this signal: Support capacity and user task quality have not been proven. This question instead requires an action that can adapt rollout controls to local obligations without lowering core safety. Availability, privacy law and employee consultation rules differ by region. Monitor regulatory changes and unapproved cross-border transfers. A feature or data practice acceptable in one jurisdiction may require different notice, consultation or deployment controls elsewhere. Review local obligations while preserving consistent minimum safeguards.
Question 500
At Evergreen Research, an executive declares an AI pilot successful because weekly users doubled. Evidence: No customer quality, time savings or cost data were collected. The business goal is to define success through business outcomes rather than activity. The project owner requires a measurable acceptance criterion before signoff. Which evidence would most convincingly demonstrate that the requirement has been met?
- Each blocker has an owner and escalation channel
- Readers can distinguish illustration from verified event evidence
- Net time savings and quality both meet the stated thresholds
- Employees can explain both benefits and guardrails
Explanation
Correct Answer
C:
No customer quality, time savings or cost data were collected. Net time savings and quality both meet the stated thresholds.
Incorrect Answers
A:
“Each blocker has an owner and escalation channel” validates an outcome in a case where the evidence is: IT, HR, security and functional leaders run disconnected activities. This question instead requires an action that can define success through business outcomes rather than activity. No customer quality, time savings or cost data were collected. Net time savings and quality both meet the stated thresholds.
B:
“Readers can distinguish illustration from verified event evidence” validates an outcome in a case where the evidence is: The image may falsely imply a real situation occurred. This question instead requires an action that can define success through business outcomes rather than activity. No customer quality, time savings or cost data were collected. Net time savings and quality both meet the stated thresholds.
D:
“Employees can explain both benefits and guardrails” validates an outcome in a case where the evidence is: Teams miss updates and do not understand why workflows change. This question instead requires an action that can define success through business outcomes rather than activity. No customer quality, time savings or cost data were collected. Net time savings and quality both meet the stated thresholds.