Question 170
An approved classification workflow handles standard requests accurately, while a small set of high-impact anomalies always needs manual judgment. What should the responsible employee choose?
- Expand a regression suite of already-successful routine cases
- Measure quality after each optimization change
- Separate human review for exceptions from routine processing
- Keep a known-good configuration for rollback
Explanation
Correct Answer
C:Exception routing limits reviewer workload without relinquishing control over uncertain cases. Criteria and monitoring keep the boundary safe.
Incorrect Answers
A:A maintained suite covering typical and edge cases detects quality drift, a useful capability for checking robustness across representative inputs but not a solution to concentrating human attention on atypical high-risk items.
B:Cost, accuracy, and completion must be measured together on the same test set; that is relevant to validating trade-offs after a cost reduction, not the stated need for concentrating human attention on atypical high-risk items.
D:Versioned instructions and rollback criteria allow recovery while the defect is investigated, while the requirement here is concentrating human attention on atypical high-risk items instead of reversing a harmful prompt release.
Question 171
A staff briefing on a regulation might be outdated because the change was announced this week and no retrieval date is specified. Which approach best satisfies this requirement?
- List analytical assumptions using last month’s regulation
- Ask for current sources and a defined timeframe for research
- Demand deliberate variety in brainstorming
- Provide audience, voice and length for a draft
Explanation
Correct Answer
B:Research instructions should define recency and require checkable citations. Current external facts need retrieval rather than unsupported memory.
Incorrect Answers
A:Asking for the underlying figures and calculation method makes the analysis reproducible and easy to recompute outside Claude; that distinction matters because the specific gap is gathering time-bounded external evidence, not making numerical analysis auditable.
C:Asking for variation across audience, channel, format, and effort broadens the idea space, yet the scenario asks for gathering time-bounded external evidence rather than exploring differentiated candidate ideas.
D:Drafting constraints guide tone, concision, and authorized commitments; a factual research method alone would not determine an appropriate customer message; that is relevant to drafting a message within communication guardrails, not the stated need for gathering time-bounded external evidence.
Question 172
A pricing comparison formula seems correct for standard orders but may fail when a discount applies only above a threshold. What is the most defensible approach?
- Test numerical claims on edge cases
- Search for conflicting policy editions
- Preserve a reference to the exact policy version
- Request exact cited extracts for review
Explanation
Correct Answer
A:Boundary inputs reveal rule implementation defects hidden by typical examples. Verification should include relevant thresholds and exception conditions.
Incorrect Answers
B:Identify the controlling revision and jurisdiction rather than arbitrarily selecting a convenient date, a useful capability for reconciling conflicting official statements but not a solution to testing a numeric rule at its boundaries.
C:Version identifiers and retrieval dates make evidence traceable; a generic link to a changing document may not reproduce the same rule later; that is relevant to preserving an auditable source version, not the stated need for testing a numeric rule at its boundaries.
D:Brief source extracts with locations focus inspection on specific evidence; this helps with making evidence passages inspectable while leaving testing a numeric rule at its boundaries unsolved.
Question 173
A team has recurring formatting rules but changes the reporting month and figures every week; the rules should persist without freezing old values. How should the team proceed?
- Refresh the source figures each week but retype the same house-format rules into every chat
- Separate reusable rules from ephemeral facts
- Use a concise carry-forward summary of a resolved task
- Keep a chat on one coherent problem
Explanation
Correct Answer
B:Persistent instructions belong in shared configuration, while current inputs remain task-specific. Mixing them invites stale conclusions.
Incorrect Answers
A:Persistent knowledge must be curated for currency; remembering outdated information well is not operational correctness; that is relevant to updating long-lived context to the current revision, not the stated need for distinguishing durable instructions from changing data.
C:A structured handoff discards irrelevant exploratory chatter while preserving actionable evidence, a useful capability for compressing history into verified work products but not a solution to distinguishing durable instructions from changing data.
D:Continuing a focused thread preserves immediate assumptions and revisions, while the requirement here is distinguishing durable instructions from changing data instead of maintaining useful same-task continuity.
Question 174
Faster report drafting has cut authoring time dramatically, but the approval queue still adds four days to each release. What is the most defensible approach?
- Establish a manual report-production fallback for outages
- Redesign the end-to-end workflow when approval delays dominate
- Retain review for consequential actions
- Augment an existing drafting step first
Explanation
Correct Answer
B:When the bottleneck moves, reconsider handoffs, exception handling, and reviewer allocation. Further drafting optimization has diminishing value.
Incorrect Answers
A:A fallback protects essential operations during outages or degraded outputs, while the requirement here is addressing a downstream constraint through redesign instead of maintaining service during tool failures.
C:Control points should be optimized, not deleted blindly; consequential output still requires authorized validation; that is relevant to preserving mandatory assurance in a new flow, not the stated need for addressing a downstream constraint through redesign.
D:Inserting Claude at the drafting stage can reduce effort while preserving approvals and handoffs, a useful capability for improving one step without changing roles but not a solution to addressing a downstream constraint through redesign.
Question 175
A workspace contains every historical approval manual ever issued, and generated procedures often merge steps from different editions. Which action would address the stated problem most directly?
- Place the recurring rule only in one transient conversation for the shared Project
- Monitor repeated failures by category
- Remove superseded material rather than accumulating layers
- Run regression tests after a new file upload
- Verify continuing connector permissions
Explanation
Correct Answer
C:Curated replacement reduces version conflict. An archive can remain elsewhere without being active reference context.
Incorrect Answers
A:Chat-level guidance does not automatically become persistent instructions for every new conversation in a Project, which does not replace the required process of preventing historical policy mixing.
B:Categorized feedback reveals systematic defects and directs targeted configuration changes, a useful capability for prioritizing fixes from recurring failure patterns but not a solution to preventing historical policy mixing.
D:Standard test prompts compare expected behaviors before and after the update, yet the scenario asks for preventing historical policy mixing rather than detecting unintended effects of knowledge updates.
E:Access and connector permissions must reflect current entitlements; workflow design should not presume historical access remains valid; that distinction matters because the specific gap is preventing historical policy mixing, not keeping live-source permissions current.
Question 176
A project proposal claims all Claude chats have identical retention and training settings regardless of account type, without inspecting the organization's contract. Which decision is the strongest fit for this task?
- Request a special policy exception before verifying the organization’s actual plan and contract
- Review policy compliance after a workflow changes
- Limit vendor promises to verified platform behavior
- Include training before a large rollout
Explanation
Correct Answer
C:Product and data-handling assurances depend on actual terms and configuration. Verify commitments instead of making universal claims.
Incorrect Answers
A:Document the business need and request review through the designated exception process, yet the scenario asks for validating organizational data-handling claims rather than seeking authorized deviation from standard tools.
B:Changed purpose and audience may change required approvals and controls; reassess governance when risk scope expands; that distinction matters because the specific gap is validating organizational data-handling claims, not updating controls when the use case evolves.
D:Governance requires user awareness of permitted tasks, data rules, escalation paths, and review obligations, a useful capability for preparing workers to follow AI policy but not a solution to validating organizational data-handling claims.
Question 177
A purchasing manager has valid vendor quotations but needs a transparent recommendation balancing unit price, delivery risk, and maintainability. What should the responsible employee choose?
- Add promotional brand examples
- Ask for unconventional alternatives before narrowing
- Ask for source-linked synthesis rather than unsupported recall
- Specify a decision framework for trade-off analysis
Explanation
Correct Answer
D:Analysis requires explicit criteria, relative importance, and assumptions so a reviewer can challenge the decision rather than accept intuition.
Incorrect Answers
A:Concrete references let the model imitate style while keeping new details original, yet the scenario asks for weighing competing decision factors rather than matching a specific communication voice.
B:Broad category prompts counter anchoring on the first obvious solution; feasibility review belongs in a later evaluation stage; this helps with widening the initial solution space while leaving weighing competing decision factors unsolved.
C:A research-and-synthesis task needs the original materials and traceable support for comparative claims, while the requirement here is weighing competing decision factors instead of comparing verified documents rather than remembered claims.
Question 178
An educational brochure generated by Claude includes medication guidance that families might treat as individualized medical advice. What is the most appropriate next action?
- Require clinician review for health-related recommendations
- Treat medication guidance like a casual internal draft requiring only light editorial review
- Require human adjudication for hiring decisions
- Require legal approval before sending binding advice
Explanation
Correct Answer
A:A qualified healthcare professional must validate medical claims and safe scope before release. The model should not be final clinical authority.
Incorrect Answers
B:A basic relevance and appropriateness review is proportionate to low-impact ideation; that distinction matters because the specific gap is obtaining clinical oversight for consequential advice, not using proportionate oversight for low-stakes ideas.
C:Employment decisions need authorized human evaluation, relevant evidence, fairness controls, and compliant processes, a useful capability for preserving accountability in personnel decisions but not a solution to obtaining clinical oversight for consequential advice.
D:Legal specialists must confirm the interpretation and approved wording; a well-phrased model output is not authorization to offer binding terms, yet the scenario asks for obtaining clinical oversight for consequential advice rather than obtaining competent legal sign-off.
Question 179
A colleague says Claude remembered a previous policy interpretation and wants to skip checking a new regulatory amendment. What is the most defensible approach?
- Capture the earlier policy interpretation
- Avoid treating memory as a source of current law
- Use narrowly scoped source excerpts when context is crowded
- Confirm a summarized instruction before discarding the original
Explanation
Correct Answer
B:Remembered context supports continuity, not authoritative currentness. Changing external rules still need fresh source checks.
Incorrect Answers
A:A handoff summary should preserve verified decisions and uncertainties, not simply copy the entire chat; that distinction matters because the specific gap is separating continuity from source authority, not transferring project state between workers.
C:Focused relevant extracts reduce distractions while preserving the governing details; that is relevant to concentrating context on pertinent evidence, not the stated need for separating continuity from source authority.
D:Check summary fidelity against the original decisions before restarting; compression can lose seemingly small but binding requirements, yet the scenario asks for separating continuity from source authority rather than validating a compressed context handoff.
Question 180
A manager reports 300 generated drafts per day as productivity even though only 40 receive approval and reach customers. What is the most appropriate next action?
- Use a structured exception queue
- Release consequential decisions based on the draft alone during workflow design
- Measure throughput after the approval step
- Train operators on changed responsibilities
- Map the full process before automating individual tasks
Explanation
Correct Answer
C:Completed accepted outputs are a more meaningful measure than raw generation. Queue capacity determines effective throughput.
Incorrect Answers
A:Exception routing separates safe routine work from cases requiring human judgment; that is relevant to escalating high-risk or ambiguous cases, not the stated need for counting usable completed work.
B:Unreviewed consequential decisions remove accountability and do not address error and bias risks; it leaves the specified need for counting usable completed work unresolved.
D:Change management must state new boundaries and accountability; a technically functioning draft system can still fail through misuse, yet the scenario asks for counting usable completed work rather than aligning staff behavior with revised roles.
E:End-to-end mapping identifies shared constraints and opportunities; local acceleration can otherwise overload the next stage; that distinction matters because the specific gap is counting usable completed work, not seeing the entire process before task selection.