Anthropic CCAR-P (Claude Certified Architect - Professional) Practice Test 30

Question 290

A reviewer repeatedly rubber-stamps model-generated high-risk summaries under a strict 10-second quota. What should the solution architect prioritize?

  1. Redesign the review workload and require meaningful evidence inspection for critical cases.
  2. Require qualified human approval before sending an adverse decision.
  3. Provide evidence-linked diffs, highlighted uncertainties and explicit approve/reject actions.
  4. Route uncertain or high-risk cases to clinicians using calibrated thresholds and workload monitoring.

Explanation

Correct Answer

A:

Nominal human presence is not an effective control if the process prevents independent judgment. Record who authorized avoidance of review automation bias and keep rejection and timeout state durable so retries cannot bypass an approval decision.

Incorrect Answers

B:

This addresses human decision ownership for adverse actions; however, the problem asks for avoidance of review automation bias. Without the safeguard to redesign the review workload and require meaningful evidence inspection for critical cases, the original failure mode is still possible.

C:

That choice would be responsive to informed review with evidence, but the reported defect is different. It leaves avoidance of review automation bias unproved because there is no commitment to redesign the review workload and require meaningful evidence inspection for critical cases.

D:

There is a legitimate use case for confidence- and severity-based review allocation, but it does not supply the test or enforcement needed for avoidance of review automation bias. The design needs to redesign the review workload and require meaningful evidence inspection for critical cases.

Question 291

A support assistant is permitted to draft refund emails but must not issue payments until a supervisor confirms details. Which design choice best matches the stated constraint?

  1. Persist rejection state and prevent automatic re-entry without new authorization.
  2. Keep dual authorization external to the model and store approval identities in an audit trail.
  3. Separate recommendation, approval and execution with authenticated actor identities.
  4. Fail closed for mandatory approvals and notify an authorized fallback reviewer.

Explanation

Correct Answer

C:

Clear duties keep model suggestions from becoming financial instructions automatically. Record who authorized segregation between drafting and committing and keep rejection and timeout state durable so retries cannot bypass an approval decision.

Incorrect Answers

A:

This recommendation targets durable human veto enforcement at a different point in the workflow. It leaves segregation between drafting and committing outside the design contract and does not require the team to separate recommendation, approval and execution with authenticated actor identities.

B:

This is a valid architectural concern when optimizing audited multi-party access approval. Here it leaves segregation between drafting and committing unaddressed, whereas the intended change must separate recommendation, approval and execution with authenticated actor identities before the case can be considered resolved.

D:

By itself, this improves approval timeout safety, but the case is not asking the architect to optimize that dimension. The requirement is segregation between drafting and committing, supported by an approach to separate recommendation, approval and execution with authenticated actor identities.

Question 292

An escalation queue times out and a model proceeds with the rejected assumption anyway. Which architecture adjustment is most appropriate?

  1. Fail closed for mandatory approvals and notify an authorized fallback reviewer.
  2. Persist rejection state and prevent automatic re-entry without new authorization.
  3. Enforce a deterministic amount-based approval gate in the workflow engine.
  4. Separate recommendation, approval and execution with authenticated actor identities.

Explanation

Correct Answer

A:

Timeout policy must preserve the gate's intent; silence cannot be treated as consent. Record who authorized approval timeout safety and keep rejection and timeout state durable so retries cannot bypass an approval decision.

Incorrect Answers

B:

For a workflow whose primary weakness was durable human veto enforcement, that proposal could be reasonable. Here the unmet obligation is approval timeout safety, and that requires the system to fail closed for mandatory approvals and notify an authorized fallback reviewer.

C:

Although rule-governed risk-based approval can be valuable, it is not the acceptance condition in the question. The missing provision is approval timeout safety, addressed by the control to fail closed for mandatory approvals and notify an authorized fallback reviewer.

D:

It is aimed at a different outcome, segregation between drafting and committing. No evidence is given that it provides approval timeout safety; the needed mechanism is to fail closed for mandatory approvals and notify an authorized fallback reviewer at the appropriate boundary.

Question 293

A clinical assistant receives conflicting drug data from two trusted databases. Which next step is most defensible?

  1. Redesign the review workload and require meaningful evidence inspection for critical cases.
  2. Require qualified human approval before sending an adverse decision.
  3. Distinguish sampled quality assurance from per-decision approval in release records.
  4. Surface the discrepancy and refer the decision to a qualified clinician rather than averaging results.

Explanation

Correct Answer

D:

An evidence conflict requires accountable domain judgment, not synthetic consensus. Record who authorized escalation of contradictory medical evidence and keep rejection and timeout state durable so retries cannot bypass an approval decision.

Incorrect Answers

A:

There is a legitimate use case for avoidance of review automation bias, but it does not supply the test or enforcement needed for escalation of contradictory medical evidence. The design needs to surface the discrepancy and refer the decision to a qualified clinician rather than averaging results.

B:

The proposed measure is aimed at human decision ownership for adverse actions, not escalation of contradictory medical evidence. The architecture must instead surface the discrepancy and refer the decision to a qualified clinician rather than averaging results; a separate change is still needed for escalation of contradictory medical evidence.

C:

The alternative could reduce risk associated with truthful evidence of human review scope, yet the described behavior requires escalation of contradictory medical evidence. That outcome depends on the capability to surface the discrepancy and refer the decision to a qualified clinician rather than averaging results.

Question 294

A company wants to automate user access grants with AI but audit rules require two-person approval. Which approach should be recommended?

  1. Separate recommendation, approval and execution with authenticated actor identities.
  2. Require qualified human approval before sending an adverse decision.
  3. Distinguish sampled quality assurance from per-decision approval in release records.
  4. Fail closed for mandatory approvals and notify an authorized fallback reviewer.
  5. Keep dual authorization external to the model and store approval identities in an audit trail.

Explanation

Correct Answer

E:

Separation of duties remains mandatory even if Claude prepares an access-change request. Record who authorized audited multi-party access approval and keep rejection and timeout state durable so retries cannot bypass an approval decision.

Incorrect Answers

A:

By itself, this improves segregation between drafting and committing, but the case is not asking the architect to optimize that dimension. The requirement is audited multi-party access approval, supported by an approach to keep dual authorization external to the model and store approval identities in an audit trail.

B:

Its main effect concerns human decision ownership for adverse actions. The case explicitly hinges on audited multi-party access approval; to resolve it reliably, the implementation has to keep dual authorization external to the model and store approval identities in an audit trail, which this choice does not establish.

C:

The option might improve truthful evidence of human review scope; that remains insufficient for audited multi-party access approval. The case is resolved by audited multi-party access approval, which depends on a design that will keep dual authorization external to the model and store approval identities in an audit trail.

D:

The proposed action prioritizes approval timeout safety. That can coexist with the right solution, but it cannot replace audited multi-party access approval: the application must still keep dual authorization external to the model and store approval identities in an audit trail.

Question 295

A bank's planned AI system processes customer identifiers in a jurisdiction with strict residency rules. Which change most directly satisfies this requirement?

  1. Perform a control-by-control gap assessment with legal and security owners.
  2. Preserve evidence, decision inputs, model and policy versions and human approval records.
  3. Map data flows across models, connectors, caches and logs before selecting approved hosting.
  4. Implement deletion propagation to derived indexes, caches and applicable retained artifacts.

Explanation

Correct Answer

C:

Regulatory boundaries attach to all copies and derived artifacts, not just the primary database. Map complete residency-aware data flow to accountable owners, data flows and retained evidence, including connectors and derived AI artifacts.

Incorrect Answers

A:

It is aimed at a different outcome, documented controls mapped to requirements. No evidence is given that it provides complete residency-aware data flow; the needed mechanism is to map data flows across models, connectors, caches and logs before selecting approved hosting at the appropriate boundary.

B:

It emphasizes traceable high-impact decision records without closing the stated gap in complete residency-aware data flow. By contrast, the remedy to map data flows across models, connectors, caches and logs before selecting approved hosting acts on the condition that makes this scenario fail.

D:

Improving complete deletion propagation does not imply complete residency-aware data flow. The selected design must map data flows across models, connectors, caches and logs before selecting approved hosting, whereas this choice omits that direct protection or verification step.

Question 296

A hospital assistant uses patient records and the development team wants to persist full prompts for unlimited debugging. Which solution best resolves this operational gap?

  1. Define response accountability, escalation triggers and evidence preservation before launch.
  2. Combine encryption with least privilege, transport protection, identity controls and data minimization.
  3. Apply source-specific lifecycle and deletion controls through the retrieval stores.
  4. Define purpose-limited retention, minimization and access policies for clinical data.

Explanation

Correct Answer

D:

Operational observability must respect confidentiality and retention obligations as part of system design. Map restricted retention of clinical content to accountable owners, data flows and retained evidence, including connectors and derived AI artifacts.

Incorrect Answers

A:

The proposed measure is aimed at governed incident response obligations, not restricted retention of clinical content. The architecture must instead define purpose-limited retention, minimization and access policies for clinical data; a separate change is still needed for restricted retention of clinical content.

B:

The relevant trade-off is not primarily multi-layer confidentiality compliance. A successful fix needs restricted retention of clinical content; otherwise restricted retention of clinical content remains unverified even after this change.

C:

The control would be justified if policy-aligned retention in search indexes were the limiting factor. In this scenario the decisive target is restricted retention of clinical content; success requires an architecture that will define purpose-limited retention, minimization and access policies for clinical data.

Question 297

A public institution must explain why an AI recommendation led to a particular adverse action. What design decision would best fit these facts?

  1. Version prompts and attach deployment identifiers to immutable audit events.
  2. Bind actor identity, authorization decision and mutation result in tamper-aware audit records.
  3. Preserve evidence, decision inputs, model and policy versions and human approval records.
  4. Apply source-specific lifecycle and deletion controls through the retrieval stores.

Explanation

Correct Answer

C:

Accountability requires reconstructing the decision context, not just saving a final paragraph. Map traceable high-impact decision records to accountable owners, data flows and retained evidence, including connectors and derived AI artifacts.

Incorrect Answers

A:

The proposed action prioritizes historical policy reproducibility. That can coexist with the right solution, but it cannot replace traceable high-impact decision records: the application must still preserve evidence, decision inputs, model and policy versions and human approval records.

B:

The design concern is real but different: attributable authorization audit evidence. It does not impose the specific requirement for traceable high-impact decision records; a suitable solution must preserve evidence, decision inputs, model and policy versions and human approval records and demonstrate the result.

D:

That choice would be responsive to policy-aligned retention in search indexes, but the reported defect is different. It leaves traceable high-impact decision records unproved because there is no commitment to preserve evidence, decision inputs, model and policy versions and human approval records.

Question 298

A legal retrieval index contains documents with varying retention schedules, but one index TTL is applied to all. Which intervention most directly addresses the failure?

  1. Implement deletion propagation to derived indexes, caches and applicable retained artifacts.
  2. Perform a control-by-control gap assessment with legal and security owners.
  3. Apply source-specific lifecycle and deletion controls through the retrieval stores.
  4. Define purpose-limited retention, minimization and access policies for clinical data.

Explanation

Correct Answer

C:

Search infrastructure must honor originating records' obligations rather than flattening them away. Map policy-aligned retention in search indexes to accountable owners, data flows and retained evidence, including connectors and derived AI artifacts.

Incorrect Answers

A:

Improving complete deletion propagation does not imply policy-aligned retention in search indexes. The selected design must apply source-specific lifecycle and deletion controls through the retrieval stores, whereas this choice omits that direct protection or verification step.

B:

This recommendation targets documented controls mapped to requirements at a different point in the workflow. It leaves policy-aligned retention in search indexes outside the design contract and does not require the team to apply source-specific lifecycle and deletion controls through the retrieval stores.

D:

This would make sense as an additional project activity for restricted retention of clinical content. The priority required here is policy-aligned retention in search indexes, so the workflow still needs a control that will apply source-specific lifecycle and deletion controls through the retrieval stores.

Question 299

A payments company plans a cross-border connector without verifying which subprocessors receive personal data. What should the solution architect prioritize?

  1. Version prompts and attach deployment identifiers to immutable audit events.
  2. Define purpose-limited retention, minimization and access policies for clinical data.
  3. Perform a control-by-control gap assessment with legal and security owners.
  4. Inventory third-party processors and assess transfer mechanisms and contractual duties.

Explanation

Correct Answer

D:

Integration architecture must include external data recipients in compliance assessment. Map governed third-party data transfers to accountable owners, data flows and retained evidence, including connectors and derived AI artifacts.

Incorrect Answers

A:

The option belongs to the broader work on historical policy reproducibility, but it cannot by itself deliver governed third-party data transfers. A decision on this case needs the team to inventory third-party processors and assess transfer mechanisms and contractual duties and verify that outcome.

B:

This shifts attention toward restricted retention of clinical content when the relevant decision is about governed third-party data transfers. Unless the architecture can inventory third-party processors and assess transfer mechanisms and contractual duties, the failure described in the stem remains unresolved.

C:

For a workflow whose primary weakness was documented controls mapped to requirements, that proposal could be reasonable. Here the unmet obligation is governed third-party data transfers, and that requires the system to inventory third-party processors and assess transfer mechanisms and contractual duties.

Question 300

An enterprise audits production agents, but tool calls are logged without the requesting employee's identity. Which design choice best matches the stated constraint?

  1. Bind actor identity, authorization decision and mutation result in tamper-aware audit records.
  2. Version prompts and attach deployment identifiers to immutable audit events.
  3. Combine encryption with least privilege, transport protection, identity controls and data minimization.
  4. Preserve evidence, decision inputs, model and policy versions and human approval records.

Explanation

Correct Answer

A:

A tool result alone cannot establish who initiated an action or whether it was permitted. Map attributable authorization audit evidence to accountable owners, data flows and retained evidence, including connectors and derived AI artifacts.

Incorrect Answers

B:

There is a legitimate use case for historical policy reproducibility, but it does not supply the test or enforcement needed for attributable authorization audit evidence. The design needs to bind actor identity, authorization decision and mutation result in tamper-aware audit records.

C:

Consider its actual operating scope: multi-layer confidentiality compliance. The requested assurance is attributable authorization audit evidence, which calls for an implementation that can bind actor identity, authorization decision and mutation result in tamper-aware audit records, not merely for a related improvement.

D:

The alternative could reduce risk associated with traceable high-impact decision records, yet the described behavior requires attributable authorization audit evidence. That outcome depends on the capability to bind actor identity, authorization decision and mutation result in tamper-aware audit records.

Claude Certified Architect - Professional (CCAR-P) Practice Tests