Question 170
An independent assessment of a telecommunications support Claude system reports that after-the-fact logs cannot reverse a completed sensitive transfer. Which change best closes the exposed control gap?
- Apply Defense-in-depth incident response
- Apply High-risk human approval gate
- Apply Output privacy filter
- Apply Rate and spend limits by tenant
Explanation
Correct Answer
B:The problem lies in the absence of a mechanism where the control layer can stop before execution and require authorized human confirmation. Reinstating it, backed by reviewer authority, timeout path and approval evidence, limits the exposure that after-the-fact logs cannot reverse a completed sensitive transfer; the engineering team should track unapproved high-risk actions prevented during simulations.
Incorrect Answers
A:This addresses the failure where a single classifier cannot handle every emergent attack pattern, because the control layer can combine prevention, detection, containment and audited recovery; but this remedy operates on another boundary, not the case where after-the-fact logs cannot reverse a completed sensitive transfer.
C:This addresses the failure where upstream input minimization alone may not catch derived disclosures, because the control layer can enforce recipient-specific redaction and validated output fields; yet it would not correct the observed weakness: after-the-fact logs cannot reverse a completed sensitive transfer.
D:Where one customer can monopolize shared model capacity through abusive retries, the control layer can enforce bounded usage budgets and anomaly controls per tenant; but this remedy operates on another boundary, not the case where after-the-fact logs cannot reverse a completed sensitive transfer.
Question 171
Following a regression in public-sector casework, the team establishes that a shared service token can reveal records across users. Which proposed redesign most directly prevents recurrence?
- Deploy Tenant-bound retrieval filtering
- Deploy OAuth consent and delegated scopes
- Deploy Per-user delegated authorization
- Deploy Authenticated webhook processing
- Deploy Server-side tool authorization
Explanation
Correct Answer
C:The remediation acts at the relevant design boundary: the integration can enforce each call using the requesting user's authorized scope. That counteracts the observed hazard that a shared service token can reveal records across users. Rollout should include propagated identity and resource-level access policy, with regression evidence from cross-tenant unauthorized-read attempts successfully denied.
Incorrect Answers
A:This addresses the failure where post-generation redaction cannot undo a cross-tenant retrieval leak, because the integration can apply tenant and document ACL filters before results reach the model; whereas the public-sector casework defect follows the path in which a shared service token can reveal records across users.
B:This addresses the failure where administrator-wide credentials bypass user-controlled consent semantics, because the integration can obtain user consent for minimum scopes and handle revocation; yet it would not correct the observed weakness: a shared service token can reveal records across users.
D:The alternative reduces the risk that trusting an inbound URL lets attackers forge operational events because the integration can verify message authenticity, timestamp and replay protection before accepting commands; however, the public-sector casework defect follows the path in which a shared service token can reveal records across users.
E:This addresses the failure where prompt rules alone cannot serve as a security boundary, because the executor evaluates policy before invoking protected downstream APIs; yet it would not correct the observed weakness: a shared service token can reveal records across users.
Question 172
A cost-cutting proposal for manufacturing quality eliminates Security policy violation rate. Which undesirable consequence would become most credible after that change?
- A failure where generic satisfaction scores may reward confident errors
- A failure where a successful business task may still leak restricted records
- A failure where general answer accuracy can hide rare severe tool abuse
- A failure where citation formatting alone does not prove support for the claim
Explanation
Correct Answer
B:This is the direct residual hazard: a successful business task may still leak restricted records. The omitted intervention matters because the evaluation process can evaluate denied and erroneously permitted requests against authoritative policy. A reviewer should demand labeled authorization test cases and measure false-allow and false-deny rates by scope before treating the risk as controlled.
Incorrect Answers
A:If executives see fluent answers but cannot judge operational correctness, generic satisfaction scores may reward confident errors; whereas removing Security policy violation rate primarily exposes the risk that a successful business task may still leak restricted records.
C:If an assistant can invoke sensitive operations through tools, general answer accuracy can hide rare severe tool abuse; however, the named safeguard addresses the distinct hazard that a successful business task may still leak restricted records.
D:When the system summarizes regulated policies with citations, citation formatting alone does not prove support for the claim; whereas removing Security policy violation rate primarily exposes the risk that a successful business task may still leak restricted records.
Question 173
During an acceptance review in contract administration, Context-capacity fit before model selection is operational but its benefit has not been measured. Auditors are concerned that nominal context limits alone do not prove reliable use of every passage. Which evidence most directly settles the question?
- Evaluate relevant-fact recall by document position and input length
- Evaluate safe fallback completion rate under simulated outages
- Evaluate audited synthesis accuracy on difficult cases
- Evaluate routed quality at the target cost and p95 latency
Explanation
Correct Answer
A:The acceptance measure focuses on relevant-fact recall by document position and input length at the task or workflow boundary. Model fluency would not establish whether the model configuration can test effective context utilization and recall before choosing a model configuration, and it could conceal cases where nominal context limits alone do not prove reliable use of every passage.
Incorrect Answers
B:Where primary inference service can time out during a critical support window, a relevant check would establish whether the model configuration can route to an approved fallback model with compatible output contracts; however, that proxy would not demonstrate relevant-fact recall by document position and input length for the selected intervention.
C:The alternative indicator checks whether a stronger model can justify tradeoffs across heterogeneous evidence when multi-document arbitration requires several interdependent inferences; but the acceptance gate must instead inspect relevant-fact recall by document position and input length to detect the targeted failure.
D:The alternative indicator checks whether a fast adequate tier handles easy cases and escalates only uncertain ones when most messages are short and low-complexity while a minority need deep analysis; by contrast, the contract administration release needs evidence from relevant-fact recall by document position and input length.
Question 174
A contract administration rollout must satisfy this design premise: staged hypotheses tie each funding tranche to measured evidence. Choose TWO linked items that form a workable architecture change and its essential preparation. Choose TWO.
- Implement Feasibility pilot with representative cases
- Track reviewer override and acceptance rates by decision category
- Establish an agreed baseline and accountable business owner
- Implement Phased value hypothesis testing
- Establish hypotheses with owners and stage-gate thresholds
Explanation
Correct Answers
D:The architecture component is justified by its behavior: staged hypotheses tie each funding tranche to measured evidence. It targets the exposure that a portfolio of speculative features can crowd out working high-value uses, while the complementary selected actions provide readiness and, for a three-part package, evidence.
E:Choosing this setup item establishes hypotheses with owners and stage-gate thresholds, which is needed for reliable operation when staged hypotheses tie each funding tranche to measured evidence. The Phased value hypothesis testing rollout should not be declared ready until hypotheses with owners and stage-gate thresholds is verified in its operating environment.
Incorrect Answers
A:This design choice is aimed at cases where the proposed automation encounters exceptions that have never been tested; it is effective because small-scope trials expose feasibility and exception handling before a broad commitment. The rollout described is instead governed by the case where a team wants to fund a large platform before validating which use case helps and requires a different primary intervention; however, the intended intervention requires that staged hypotheses tie each funding tranche to measured evidence.
B:This test would validate a different benefit: Claude drafts ranked recommendations while a designated reviewer authorizes each action; but that indicator does not verify cost per accepted result against a defined business target for the selected architecture.
C:The alternative setup enables a different mechanism: a measurable business outcome narrows the solution boundary before model selection; whereas this intervention requires hypotheses with owners and stage-gate thresholds as its operating precondition.
Question 175
A public-sector project proposes bypassing the Use-case prioritization workshop and choosing its first Claude workflow solely on sponsor enthusiasm. Which selection risk is least controlled?
- The risk that unclear authority causes late redesign and unowned risk decisions
- The risk that feature completeness does not demonstrate production acceptability
- The risk that open-ended satisfaction language cannot determine release readiness
- The risk that political priority alone may select a low-return high-risk deployment
Explanation
Correct Answer
D:The risk is not a generic model inconvenience: political priority alone may select a low-return high-risk deployment. The proposed safeguard changes that boundary because the delivery team can jointly rank candidates by value, feasibility, harm and data readiness, and the release would need scoring rubric and representatives from delivery and risk teams plus agreed priority scores and expected benefits for shortlisted uses to show the risk remains contained.
Incorrect Answers
A:If product, security and legal teams disagree on who can approve release, unclear authority causes late redesign and unowned risk decisions; whereas removing Use-case prioritization workshop primarily exposes the risk that political priority alone may select a low-return high-risk deployment.
B:The other exposure arises when a proposal lists features but omits response time: recovery requirements and feature completeness does not demonstrate production acceptability. Omitting the requested control instead leaves the more immediate risk that political priority alone may select a low-return high-risk deployment; but this is not the immediate loss associated with omitting Use-case prioritization workshop: political priority alone may select a low-return high-risk deployment.
C:The competing hazard is that open-ended satisfaction language cannot determine release readiness in a workload where stakeholders ask for a system that is accurate enough without specifying thresholds; however, the named safeguard addresses the distinct hazard that political priority alone may select a low-return high-risk deployment.
Question 176
The telecommunications support architect proposes omitting High-risk human approval gate from the production design. What is the most direct failure mode that decision would expose?
- An exposure in which model self-restraint cannot guarantee that a bad action is blocked
- An exposure in which a single classifier cannot handle every emergent attack pattern
- An exposure in which upstream input minimization alone may not catch derived disclosures
- An exposure in which after-the-fact logs cannot reverse a completed sensitive transfer
Explanation
Correct Answer
D:Removing the safeguard exposes a failure in which after-the-fact logs cannot reverse a completed sensitive transfer. The architecture normally counters this because the control layer can stop before execution and require authorized human confirmation. A compensating design would require reviewer authority, timeout path and approval evidence and evidence from unapproved high-risk actions prevented during simulations.
Incorrect Answers
A:The other exposure arises when assistants propose privileged write operations in customer systems: model self-restraint cannot guarantee that a bad action is blocked. Omitting the requested control instead leaves the more immediate risk that after-the-fact logs cannot reverse a completed sensitive transfer; but this is not the immediate loss associated with omitting High-risk human approval gate: after-the-fact logs cannot reverse a completed sensitive transfer.
B:The competing hazard is that a single classifier cannot handle every emergent attack pattern in a workload where failures may bypass one input or model safety layer; however, the named safeguard addresses the distinct hazard that after-the-fact logs cannot reverse a completed sensitive transfer.
C:When generated summaries may echo sensitive identifiers unnecessary for recipients, upstream input minimization alone may not catch derived disclosures; whereas removing High-risk human approval gate primarily exposes the risk that after-the-fact logs cannot reverse a completed sensitive transfer.
Question 177
A lean redesign in research administration would retire Per-user delegated authorization while leaving the rest of the service unchanged. Which operational exposure is most directly increased?
- The danger that post-generation redaction cannot undo a cross-tenant retrieval leak
- The danger that administrator-wide credentials bypass user-controlled consent semantics
- The danger that a shared service token can reveal records across users
- The danger that prompt rules alone cannot serve as a security boundary
Explanation
Correct Answer
C:This is the direct residual hazard: a shared service token can reveal records across users. The omitted intervention matters because the integration can enforce each call using the requesting user's authorized scope. A reviewer should demand propagated identity and resource-level access policy and measure cross-tenant unauthorized-read attempts successfully denied before treating the risk as controlled.
Incorrect Answers
A:When one vector index holds embeddings for several client organizations, post-generation redaction cannot undo a cross-tenant retrieval leak; however, the named safeguard addresses the distinct hazard that a shared service token can reveal records across users.
B:When employees connect personal enterprise resources through an integration, administrator-wide credentials bypass user-controlled consent semantics; whereas removing Per-user delegated authorization primarily exposes the risk that a shared service token can reveal records across users.
D:The competing hazard is that prompt rules alone cannot serve as a security boundary in a workload where a model can propose actions but must not decide its own privileges; however, the named safeguard addresses the distinct hazard that a shared service token can reveal records across users.
Question 178
The defining problem for a healthcare operations pilot is that incomplete evidence should yield escalation rather than invented certainty. Which proposed approach addresses it most precisely?
- Introduce Tail-latency measurement
- Introduce Calibration and abstention quality
- Introduce Task-level acceptance rubric
- Introduce Security policy violation rate
Explanation
Correct Answer
B:The architectural distinction is that the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels. The decision should be implemented with gold labels including unanswerable requests and evaluated against risk-coverage curve and improper certainty rate; otherwise, accuracy on answered cases alone can reward refusing everything.
Incorrect Answers
A:When interactive sessions frequently exceed their response deadline, the evaluation process can track p95 and p99 end-to-end latency by request class; it does not ensure the required behavior: the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels.
C:When executives see fluent answers but cannot judge operational correctness, the evaluation process can define observable criteria for a usable output rather than aesthetic fluency; the alternative leaves the central requirement unproven: the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels.
D:For a workload where agent integration must respect access restrictions for multiple tenants, the evaluation process can evaluate denied and erroneously permitted requests against authoritative policy; it does not ensure the required behavior: the evaluation process can evaluate correctness jointly with justified abstentions across uncertainty levels.
Question 179
The education services release team needs a numerical gate for Reproducible development environment. Rather than equating deployment with success, it wants to catch cases in which ambient local settings conceal incompatibilities in shared builds. Which indicator should it track?
- Monitor agent-generated changes passing clean-environment builds
- Monitor approved connector use and access-policy compliance
- Monitor standards adherence across fresh developer workspaces
- Monitor cross-workspace access violations in test cases
Explanation
Correct Answer
A:Tracking agent-generated changes passing clean-environment builds provides direct evidence that the system delivers its intended effect: the engineering workflow can pin dependencies, tooling and environment assumptions across contributors. The measure is more informative than prompt counts because the failure being measured is that ambient local settings conceal incompatibilities in shared builds.
Incorrect Answers
B:The alternative indicator checks whether the engineering workflow can register reviewed MCP servers with clear ownership and allowed scopes when development teams connect tools with overlapping credentials and contracts; however, that proxy would not demonstrate agent-generated changes passing clean-environment builds for the selected intervention.
C:The proposed measure would suit a setting where developers get inconsistent coding behavior across branches and members, since the engineering workflow can version shared Claude Code guidance alongside the project and review changes; but the acceptance gate must instead inspect agent-generated changes passing clean-environment builds to detect the targeted failure.
D:When confidential projects share agent context, local tool access inadvertently and the engineering workflow can isolate project roots, secrets and tool permissions by approved workspace; by contrast, the education services release needs evidence from agent-generated changes passing clean-environment builds.
Question 180
During solution discovery in legal operations, candidate use cases differ in savings, uncertainty and personal-data exposure. Which design response addresses the stated constraint most directly?
- Introduce Outcome-first use-case framing
- Introduce Value-to-risk prioritization
- Introduce Human-approved recommendation service
- Introduce Task-scope exclusion for unsuitable decisions
Explanation
Correct Answer
B:The mechanism is specific: a scored decision framework surfaces high-value lower-risk starting points. A credible implementation requires a joint rubric agreed with security and operations, while its acceptance test monitors risk-adjusted impact and implementation effort per use case. The alternative approaches leave the risk that prioritizing solely by raw request volume ignores potential harm.
Incorrect Answers
A:If stakeholders disagree whether faster drafting or fewer escalations is the goal, a measurable business outcome narrows the solution boundary before model selection; that does not provide the capability actually needed: a scored decision framework surfaces high-value lower-risk starting points.
C:If caseworkers must retain authority for consequential final decisions, Claude drafts ranked recommendations while a designated reviewer authorizes each action; whereas the legal operations case must guard against the outcome that prioritizing solely by raw request volume ignores potential harm.
D:When a client demands a guarantee of fully accurate autonomous legal advice, identifying out-of-scope judgments prevents unsupported automation promises; that does not provide the capability actually needed: a scored decision framework surfaces high-value lower-risk starting points.