Microsoft AB-100 (Agentic AI Business Solutions Architect) Practice Test 69

Question 680

A planned AI solution must support this situation: Users use novel requests outside the supported topic inventory. Its prototype fails because the agent repeats an unhelpful default message. Which design best fits the constraint?

  1. Use a prompt action for bounded text transformation
  2. Use explicit topics for a tightly scripted compliance dialogue
  3. Design a fallback topic that escalates genuine nonmatches
  4. Use an agent flow for repeatable ordered business steps

Explanation

Correct Answer

C:

Users use novel requests outside the supported topic inventory; its failing behavior is that the agent repeats an unhelpful default message; applying the underlying design principle—fallback handles unmatched intent and can ask for clarification or escalate—connects the requirement to an implementable control.

Incorrect Answers

A:

For another type of architecture, prompt actions encapsulate reusable inputs and expected outputs is useful; this design must instead support users use novel requests outside the supported topic inventory despite the problem where the agent repeats an unhelpful default message.

B:

Deterministic topic logic controls known prompts and branching; its main purpose is a regulated process requires fixed questions in exact sequence, not correction of the reported defect where the agent repeats an unhelpful default message in the target process.

D:

Defined flow steps make sequencing and error paths explicit; its main purpose is approval processing requires sequential verification and notification, not correction of the reported defect where the agent repeats an unhelpful default message in the target process.

Question 681

Release reviewers will not accept a generic success message. The business scenario is that finance users need current operational data in contextual help, and the historical failure is: A static knowledge upload gives stale purchase-order states. Which test is appropriately targeted?

  1. Modify a test purchase order and validate authorized grounded retrieval
  2. Ask users to correct and accept a suggested field
  3. Try the action with a user lacking record access
  4. Draft replies for differing contract tiers and inspect accuracy

Explanation

Correct Answer

A:

The historical failure is that a static knowledge upload gives stale purchase-order states; for the setting where finance users need current operational data in contextual help, modifying a test purchase order and validating authorized grounded retrieval verifies the dependency where authorized virtual entities can surface current finance and operations data to supported agents.

Incorrect Answers

B:

The suggested test serves staff reject suggestions because approval controls are unclear, in which users cannot tell whether a value was generated or saved; the present scenario instead concerns the reported problem where a static knowledge upload gives stale purchase-order states.

C:

Although the exercise might catch hidden fields from the UI appear in the prompt payload, it would not directly validate the mechanism that authorized virtual entities can surface current finance and operations data to supported agents in the target implementation.

D:

This check exercises a different system behavior: contextual grounding provides relevant policy and customer facts; a pass there cannot establish that a static knowledge upload gives stale purchase-order states no longer occurs in the required workflow.

Question 682

After deployment, stakeholders need an interpretable measure for this use case: Debugging logs include prompts containing customer data. The pilot showed that historical traces remain available longer than approved retention. Which metric captures the needed result?

  1. Tail latency by model, retrieval and tool phase
  2. Third-party agents completing security review before activation
  3. Sensitive telemetry records retained beyond policy limit
  4. Data-policy violations blocked in integration tests

Explanation

Correct Answer

C:

This scorecard needs an outcome related to debugging logs include prompts containing customer data; sensitive telemetry records retained beyond policy limit. That indicator can be checked against the earlier symptom that historical traces remain available longer than approved retention.

Incorrect Answers

A:

That scorecard concerns the activity in which agents regularly exceed an interactive response-time objective and its result of tail latency by model, retrieval and tool phase; the reported shortcoming here is historical traces remain available longer than approved retention.

B:

The proposed metric concerns employees install external agents with unknown data pathways, whereas the failure in this case is that historical traces remain available longer than approved retention; the indicator needed here is sensitive telemetry records retained beyond policy limit.

D:

The proposed metric concerns a Copilot Studio agent can read corporate data and use public messaging, whereas the failure in this case is that historical traces remain available longer than approved retention; the indicator needed here is sensitive telemetry records retained beyond policy limit.

Question 683

The architect wants evidence before altering the deployment. In a workflow where users must remain in a proprietary work-order application, requiring a separate chat portal increases context switching. What should the team examine first?

  1. Compare feature gaps and vendor operating obligations
  2. Check whether the required API can be surfaced through supported actions
  3. Capture advertised capabilities and actual authorized tool calls
  4. Observe navigation abandonment and contextual data transfer

Explanation

Correct Answer

D:

For the situation where users must remain in a proprietary work-order application, the defect is that requiring a separate chat portal increases context switching; observing navigation abandonment and contextual data transfer helps verify the cause-and-effect path: app-hosted integration brings relevant AI actions into an owned user interface.

Incorrect Answers

A:

The proposed inspection is associated with the requirement is common and product coverage is adequate, where managed functionality shifts routine platform operations to the service; it would not expose the failure mode that requiring a separate chat portal increases context switching with the required specificity.

B:

The proposed inspection is associated with users need familiar Copilot behavior plus an internal status lookup, where supported extensibility adds an action while retaining the host's orchestration; it would not expose the failure mode that requiring a separate chat portal increases context switching with the required specificity.

C:

For the failure that every agent reimplements the same proprietary tool signatures, capturing advertised capabilities and actual authorized tool calls is relevant; this question instead reports that requiring a separate chat portal increases context switching, so different evidence is needed.

Question 684

The architecture board has evidence that a shared connection identity grants sandbox agents production write access. The affected situation is: Test automation must not reach live customer records. What change most directly establishes the missing behavior?

  1. Validate agent publishing and channel configuration after promotion
  2. Use managed identities for service-to-service agent components
  3. Use managed solutions for governed downstream environments
  4. Deploy actions with compatible rollback behavior
  5. Use separate service principals and least-privilege access per environment

Explanation

Correct Answer

E:

The mechanism that environment-scoped identities isolate access and simplify incident containment fits the scenario where test automation must not reach live customer records; it addresses the defect that a shared connection identity grants sandbox agents production write access.

Incorrect Answers

A:

When solution import succeeds but users still see the previous behavior, post-import validation includes publishing and channel behavior; in the stated workflow, however, a shared connection identity grants sandbox agents production write access; the relevant architectural behavior is that environment-scoped identities isolate access and simplify incident containment.

B:

This design would make sense if a backend agent must call an approved Azure service without embedded secrets, particularly where long-lived API keys leak from configuration and diagnostics; the requested architecture instead needs the dependency that environment-scoped identities isolate access and simplify incident containment because a shared connection identity grants sandbox agents production write access.

C:

The technical behavior of this alternative is that managed solution deployment supports controlled downstream customization boundaries; the user case being assessed is test automation must not reach live customer records with a different failure: a shared connection identity grants sandbox agents production write access.

D:

This design would make sense if a new action version changes how cases are updated, particularly where rolling back the agent leaves an incompatible backend action; the requested architecture instead needs the dependency that environment-scoped identities isolate access and simplify incident containment because a shared connection identity grants sandbox agents production write access.

Question 685

The rollout supports this workflow: Customer-facing brands have several names for the same product. Leaders want a metric that responds to the reported shortcoming that chat responses mix replacement policies between product lines. Which measure best reflects success?

  1. Opportunity changes mapped to the intended business field
  2. Correct external-to-Dynamics account match rate
  3. Precision at top results for identifier and natural-language queries
  4. Wrong-product identification in service conversations

Explanation

Correct Answer

D:

For the use case where customer-facing brands have several names for the same product, the indicator wrong-product identification in service conversations is worth baselining against the earlier issue where chat responses mix replacement policies between product lines.

Incorrect Answers

A:

The alternative tracks opportunity changes mapped to the intended business field, arising from the mechanism where canonical business terminology maps language to underlying Dynamics records; its movement alone says little about the failure that chat responses mix replacement policies between product lines.

B:

For the situation where two systems share company names but not primary keys, correct external-to-Dynamics account match rate is a sensible indicator; the problem that chat responses mix replacement policies between product lines requires measurement of wrong-product identification in service conversations.

C:

Precision at top results for identifier and natural-language queries; that indicator describes a distinct workflow where product catalog queries mix synonyms and exact serial codes; it does not quantify wrong-product identification in service conversations, which is needed for the stated case.

Question 686

In a business situation where benchmarks cover only cooperative user wording, a previous build showed this defect: A malicious instruction triggers a privileged tool. Which reproducible test belongs in the release gate?

  1. Try injection, invalid inputs and out-of-scope actions
  2. Send concurrent equivalent events and inspect resulting writes
  3. Reproduce a sanitized failure and confirm the new regression catches it
  4. Ask for a privileged record using an unprivileged test identity

Explanation

Correct Answer

A:

The check is targeted rather than generic: negative tests expose policy and validation boundaries; its pass/fail evidence applies to the workflow where benchmarks cover only cooperative user wording and to the prior defect that a malicious instruction triggers a privileged tool.

Incorrect Answers

B:

The tested behavior is that concurrency evaluation tests idempotency and serialization; the necessary evidence here concerns negative tests expose policy and validation boundaries and the previously observed a malicious instruction triggers a privileged tool.

C:

This check exercises a different system behavior: feedback-derived examples convert incidents into lasting safeguards; a pass there cannot establish that a malicious instruction triggers a privileged tool no longer occurs in the required workflow.

D:

In the alternate setting where a user asks the agent to read an unrelated executive record, the test exercises authorization evaluations verify server-side deny behavior; even a pass would not show that a malicious instruction triggers a privileged tool has been prevented.

Question 687

Before introducing a fix, the team must explain why API trusts the model's text assertion of permission. The affected setting is: Prompt wording persuades the agent to submit a privileged operation. What should be inspected?

  1. Compare answers with retrieved and deliberately missing documents
  2. Compare requested export destinations against policy allowlists
  3. Review backend authorization logs against actor identities
  4. Analyze failures by paraphrase group and boundary type

Explanation

Correct Answer

C:

The investigation directly tests the causal connection: server-side access checks enforce real actor entitlements; it collects evidence for a workflow where prompt wording persuades the agent to submit a privileged operation but the outcome is instead that API trusts the model's text assertion of permission.

Incorrect Answers

A:

The proposed probe concerns policy answers must be tied to a current source, where evidence and refusal evaluations enforce source-supported claims; this incident instead requires evidence about server-side access checks enforce real actor entitlements and the failure that API trusts the model's text assertion of permission.

B:

Model-level refusal alone may not prevent a tool call; examining that other path does not determine whether server-side access checks enforce real actor entitlements, which is what distinguishes the issue in the stated use case.

D:

In the process where a rewritten prompt improves a single demonstration, the diagnostic procedure could explain slightly different wording breaks correct entity selection; this particular incident instead turns on whether server-side access checks enforce real actor entitlements and why API trusts the model's text assertion of permission.

Question 688

Release reviewers will not accept a generic success message. The business scenario is that an agent will evolve continuously after its initial project delivery, and the historical failure is: The pilot has no team for monitoring, incidents, or iteration. Which test is appropriately targeted?

  1. Benchmark a simple-task holdout on both model tiers
  2. Simulate a production regression and locate the accountable response team
  3. Simulate endpoint failure and verify continuity without policy bypass
  4. Trace a representative case through each owned component

Explanation

Correct Answer

B:

The pilot has no team for monitoring, incidents, or iteration; a release test must establish that this no longer occurs; simulating a production regression and locating the accountable response team measures the specific behavior for which long-lived product ownership funds ongoing evaluation and operations.

Incorrect Answers

A:

In the alternate setting where many calls require only a small fixed-label output, the test exercises task-aware routing chooses economical models after quality gates; even a pass would not show that the pilot has no team for monitoring, incidents, or iteration has been prevented.

C:

In the alternate setting where peak load can saturate a primary model endpoint, the test exercises health-aware fallback uses a compatible approved model when needed; even a pass would not show that the pilot has no team for monitoring, incidents, or iteration has been prevented.

D:

In the alternate setting where a business process mixes deterministic calculations and open-ended judgments, the test exercises task decomposition separates deterministic services from bounded agent reasoning; even a pass would not show that the pilot has no team for monitoring, incidents, or iteration has been prevented.

Question 689

An enterprise team is making a platform design decision. Business context: Approval processing requires sequential verification and notification. Present failure: An unconstrained planner sends a notification before authorization. Which approach resolves the mismatch?

  1. Use a prompt action for bounded text transformation
  2. Use an agent flow for repeatable ordered business steps
  3. Build a code-first generative page with explicitly bound data
  4. Use a structured response contract for downstream consumption

Explanation

Correct Answer

B:

The mechanism that defined flow steps make sequencing and error paths explicit fits the scenario where approval processing requires sequential verification and notification; it addresses the defect that an unconstrained planner sends a notification before authorization.

Incorrect Answers

A:

When an agent must turn service notes into a fixed JSON summary, prompt actions encapsulate reusable inputs and expected outputs; in the stated workflow, however, an unconstrained planner sends a notification before authorization; the relevant architectural behavior is that defined flow steps make sequencing and error paths explicit.

C:

When staff need a contextual workspace assembled around one customer case, generated interface components can be grounded in governed app data and defined actions; in the stated workflow, however, an unconstrained planner sends a notification before authorization; the relevant architectural behavior is that defined flow steps make sequencing and error paths explicit.

D:

This design would make sense if a summarizer sends output into a ticket ingestion service, particularly where free-form paragraphs break required schema parsing; the requested architecture instead needs the dependency that defined flow steps make sequencing and error paths explicit because an unconstrained planner sends a notification before authorization.

Question 690

Security and business owners want to keep this capability: The same agent must move from development to test and production. They must also prevent the following: A maker might import or publish an agent without release approval. Which guardrail should the design incorporate?

  1. Require a tested human escalation path for unmatched cases
  2. Stop release on unresolved breaking contract mismatches
  3. Enforce approval before production import and publication
  4. Keep prior compatible API routes during release transition

Explanation

Correct Answer

C:

The business setting is the same agent must move from development to test and production, with the exposure that a maker might import or publish an agent without release approval; the enforceable protection must correspond to the path where solution-based ALM transports supported components under controlled releases.

Incorrect Answers

A:

The alternate boundary mitigates a new support agent might route urgent cases to a general queue, which is relevant to a support agent changes how it classifies priority intents; it leaves the current workflow vulnerable to a maker might import or publish an agent without release approval.

B:

The alternate boundary mitigates a breaking connector schema might be released despite failed contract tests, which is relevant to an external API changes a required field used by an agent; it leaves the current workflow vulnerable to a maker might import or publish an agent without release approval.

D:

For the use case where a new action version changes how cases are updated, the control addresses the threat that a forced API cutover might make rollback impossible for an old agent version; the specified agent must be protected against a maker might import or publish an agent without release approval.

Agentic AI Business Solutions Architect (AB-100) Practice Tests