Pass IAPP AIGP Exam in First Attempt Guaranteed!
Get 100% Latest Exam Questions, Accurate & Verified Answers!
30 Days Free Updates, Instant Download!
IAPP AIGP Practice Test 36
Question 350
A public-service translation model is trained mainly on urban speech and performs poorly for rural dialect speakers. Which governance reasoning best supports the appropriate response?
Explanation
Correct Answer
Training distribution should represent the deployment population rather than the easiest data source. This reasoning fits representative sampling for rural users because the observed gap is that dataset profiling reveals that rural utterances represent a tiny share of the intended population. It points to the practical intervention: Stratify and validate training and test samples for the actual served language communities. The outcome should then show that evaluation demonstrates acceptable subgroup coverage and performance against defined criteria.
Incorrect Answers
Reproducibility turns an evaluation assertion into testable evidence. This reasoning fits representative sampling for rural users because the observed gap is that dataset profiling reveals that rural utterances represent a tiny share of the intended population. Training distribution should represent the deployment population rather than the easiest data source.
Data integrity checks are essential governance controls for model training and retraining. This reasoning fits representative sampling for rural users because the observed gap is that dataset profiling reveals that rural utterances represent a tiny share of the intended population. Training distribution should represent the deployment population rather than the easiest data source.
Synthetic generation does not by itself guarantee privacy or lawful disclosure. This reasoning fits representative sampling for rural users because the observed gap is that dataset profiling reveals that rural utterances represent a tiny share of the intended population. Training distribution should represent the deployment population rather than the easiest data source.
Question 351
An HR department signs a résumé-screening subscription without a documented bias or purpose assessment. Which test result would best demonstrate that the actual governance concern has been addressed?
Explanation
Correct Answer
HR and governance officers document release criteria and mitigation owners before deployment. The required intervention is to use a procurement gate covering intended use, discrimination risks, provider terms and human controls. The particular diagnostic issue is that purchase approval contains cost forecasts but no evidence of model behavior or fairness review. The governing distinction is that aI procurement is an organizational risk decision, not merely a software purchase.
Incorrect Answers
Every customer-facing experiment has a signed risk record and constrained pilot plan before access. AI procurement is an organizational risk decision, not merely a software purchase. HR and governance officers document release criteria and mitigation owners before deployment.
Sample workflows demonstrate confidential designs remain inside approved processing boundaries. AI procurement is an organizational risk decision, not merely a software purchase. HR and governance officers document release criteria and mitigation owners before deployment.
Material updates are tied to before-and-after evaluations and accountable approval. AI procurement is an organizational risk decision, not merely a software purchase. HR and governance officers document release criteria and mitigation owners before deployment.
Question 352
A retailer uses an applicant-ranking system and sees higher reject rates for equally qualified candidates from one protected group. Which governance reasoning best supports the appropriate response?
Explanation
Correct Answer
AI-assisted employment decisions remain subject to discrimination rules even when scores are automated. This reasoning fits discrimination risk in hiring because the observed gap is that audits show unexplained group disparities unrelated to job-relevant criteria. It points to the practical intervention: Assess applicable employment nondiscrimination duties and validate the model for unlawful disparate outcomes. The outcome should then show that a documented remediation and qualified legal review address the adverse-impact findings before use.
Incorrect Answers
Indirect features may reproduce prohibited discrimination even without explicit protected traits. AI-assisted employment decisions remain subject to discrimination rules even when scores are automated.
Trade-secret protections can be compromised by unmanaged disclosure to third parties. AI-assisted employment decisions remain subject to discrimination rules even when scores are automated.
Public accessibility does not necessarily grant a license for every AI training use. AI-assisted employment decisions remain subject to discrimination rules even when scores are automated.
Question 353
A public housing service wants AI to prioritize applications but has not decided whether recommendations or final automated decisions are permitted. Which governance intervention most directly addresses this specific situation?
Explanation
Correct Answer
Set a defined human decision boundary matching applicable policy and risk tolerance. The action addresses decision support versus autonomous decision rather than treating every AI control as interchangeable. The decisive signal to verify is that the proposed workflow gives an AI ranking direct authority to deny an applicant. This matters because the scope of AI decision authority materially changes governance and legal risk. Success should be measured by confirming that no rejection occurs without required human review and a documented basis for the decision.
Incorrect Answers
Compare an existing governed service with in-house development on requirements, risk, support and total cost. The scope of AI decision authority materially changes governance and legal risk. Set a defined human decision boundary matching applicable policy and risk tolerance.
Evaluate a model supporting the necessary image input and validate safety before adoption. The scope of AI decision authority materially changes governance and legal risk. Set a defined human decision boundary matching applicable policy and risk tolerance.
Assess licenses, support, security, update controls, data use terms and governance before choosing. The scope of AI decision authority materially changes governance and legal risk. Set a defined human decision boundary matching applicable policy and risk tolerance.
Question 354
An insurer discovers the newest claims model causes large unexplained rejection spikes but no tested previous version is preserved. Which governance intervention most directly addresses this specific situation?
Explanation
Correct Answer
Maintain validated rollback artifacts and a rehearsed controlled reversion procedure. The action addresses rollback of harmful version rather than treating every AI control as interchangeable. The decisive signal to verify is that operations has only the latest model image and no prior compatible feature schema. This matters because safe AI release management includes the ability to reverse harmful changes. Success should be measured by confirming that a drill restores a previously approved model and preserves audit evidence without data loss.
Incorrect Answers
Apply a formal release gate with tolerances, remediation requirements and stop authority. Safe AI release management includes the ability to reverse harmful changes. Maintain validated rollback artifacts and a rehearsed controlled reversion procedure.
Schedule evidence-based audits of model quality, validity and safety assumptions. Safe AI release management includes the ability to reverse harmful changes. Maintain validated rollback artifacts and a rehearsed controlled reversion procedure.
Require release notes, version pinning where supported and customer regression tests for provider changes. Safe AI release management includes the ability to reverse harmful changes. Maintain validated rollback artifacts and a rehearsed controlled reversion procedure.
Question 355
A housing agency has a fairness standard but no plan to disable its eligibility model when a harmful drift alert occurs. Select the ONE intervention-and-verification pair that directly resolves this specific governance gap.
Explanation
Correct Answer
Intervention: Write response thresholds, emergency suspension roles and fallback service procedures. / Verification: A tabletop exercise pauses the model and moves cases to an accountable manual workflow. The reason to require this combination is that an effective AI policy must be actionable when risks exceed tolerance.
Incorrect Answers
Intervention: Write response thresholds, emergency suspension roles and fallback service procedures. / Verification: Material updates are tied to before-and-after evaluations and accountable approval. The reason to require this combination is that an effective AI policy must be actionable when risks exceed tolerance. Intervention: Write response thresholds, emergency suspension roles and fallback service procedures. / Verification: A tabletop exercise pauses the model and moves cases to an accountable manual workflow.
Intervention: Create an AI acceptable-use policy describing allowed data, tool approvals and professional review. / Verification: A tabletop exercise pauses the model and moves cases to an accountable manual workflow. The reason to require this combination is that an effective AI policy must be actionable when risks exceed tolerance. Intervention: Write response thresholds, emergency suspension roles and fallback service procedures. / Verification: A tabletop exercise pauses the model and moves cases to an accountable manual workflow.
Intervention: Add approved-subprocessor disclosure, flow-down controls and change notification to procurement processes. / Verification: Subcontractor changes trigger documented risk review and compliant data-handling checks. The reason to require this combination is that an effective AI policy must be actionable when risks exceed tolerance. Intervention: Write response thresholds, emergency suspension roles and fallback service procedures. / Verification: A tabletop exercise pauses the model and moves cases to an accountable manual workflow.
Question 356
A regional call center buys an AI triage assistant but agents have no training on when to disregard its recommendations. Select the ONE intervention-and-verification pair that directly resolves this specific governance gap.
Explanation
Correct Answer
Intervention: Assess staff readiness, operating procedures and override skills before launch. / Verification: Trained staff correctly override unsupported advice in scenario tests and record exceptions. The reason to require this combination is that operational adoption requires prepared human users as well as working software.
Incorrect Answers
Intervention: Set a defined human decision boundary matching applicable policy and risk tolerance. / Verification: Trained staff correctly override unsupported advice in scenario tests and record exceptions. The reason to require this combination is that operational adoption requires prepared human users as well as working software. Intervention: Assess staff readiness, operating procedures and override skills before launch. / Verification: Trained staff correctly override unsupported advice in scenario tests and record exceptions.
Intervention: Select an appropriate predictive time-series model rather than defaulting to generative text output. / Verification: Validated predictions support maintenance planning with measurable false-alarm and miss rates. The reason to require this combination is that operational adoption requires prepared human users as well as working software. Intervention: Assess staff readiness, operating procedures and override skills before launch. / Verification: Trained staff correctly override unsupported advice in scenario tests and record exceptions.
Intervention: Assess staff readiness, operating procedures and override skills before launch. / Verification: Tests prove purchase execution requires an authorized human decision and policy checks. The reason to require this combination is that operational adoption requires prepared human users as well as working software. Intervention: Assess staff readiness, operating procedures and override skills before launch. / Verification: Trained staff correctly override unsupported advice in scenario tests and record exceptions.
Question 357
A public authority buys hosted inference with a contractual EU data-region promise but cannot verify where support logs go. Which governance reasoning best supports the appropriate response?
Explanation
Correct Answer
Data residency must cover operational logs and subcontractors, not only primary inference. This reasoning fits regional processing and residency terms because the observed gap is that application inference stays in-region while debug traces with personal data are exported elsewhere. It points to the practical intervention: Map all data flows including support and telemetry, then reconcile them with residency and transfer terms. The outcome should then show that observed routing and supplier records satisfy the approved processing boundaries.
Incorrect Answers
An API subscription does not automatically resolve generated-content ownership and liability. This reasoning fits regional processing and residency terms because the observed gap is that application inference stays in-region while debug traces with personal data are exported elsewhere. Data residency must cover operational logs and subcontractors, not only primary inference.
Building AI in-house can increase direct accountability rather than eliminate risk. This reasoning fits regional processing and residency terms because the observed gap is that application inference stays in-region while debug traces with personal data are exported elsewhere. Data residency must cover operational logs and subcontractors, not only primary inference.
Contractual change control is essential when behavior can shift without local code changes. This reasoning fits regional processing and residency terms because the observed gap is that application inference stays in-region while debug traces with personal data are exported elsewhere. Data residency must cover operational logs and subcontractors, not only primary inference.
Question 358
A call-center AI agent sends abusive replies while support engineers argue that only cybersecurity incidents warrant escalation. Which governance reasoning best supports the appropriate response?
Explanation
Correct Answer
AI incidents can be safety, fairness or trust failures without fitting conventional security categories. This reasoning fits escalation route for model incidents because the observed gap is that tickets for harmful outputs remain open because service desk, safety and legal teams use incompatible definitions.
Incorrect Answers
Roles in an AI supply chain are not interchangeable; each controls a different part of the risk. This reasoning fits escalation route for model incidents because the observed gap is that tickets for harmful outputs remain open because service desk, safety and legal teams use incompatible definitions. AI incidents can be safety, fairness or trust failures without fitting conventional security categories.
A policy without practical role-based understanding will not consistently change behavior. This reasoning fits escalation route for model incidents because the observed gap is that tickets for harmful outputs remain open because service desk, safety and legal teams use incompatible definitions. AI incidents can be safety, fairness or trust failures without fitting conventional security categories.
Effective AI oversight needs assigned responsibility rather than a committee name without authority. This reasoning fits escalation route for model incidents because the observed gap is that tickets for harmful outputs remain open because service desk, safety and legal teams use incompatible definitions. AI incidents can be safety, fairness or trust failures without fitting conventional security categories.
Question 359
An AI system provider stores only a product brochure and cannot show the data, evaluations and change history behind a regulated deployment. Which additional observation would most strongly confirm the particular risk or control gap?
Explanation
Correct Answer
An auditor cannot identify the tested version or its risk-control evidence. This is the diagnostic observation that distinguishes technical documentation and logging from other concerns within understand the main elements of ai-specific laws. It supports the targeted response: Prepare required technical documentation, lifecycle records and traceable logs for the applicable role. The underlying governance reason is that governance documentation must substantiate controls rather than merely restate marketing claims. A meaningful follow-up would be to verify that the compliance file links model design, evaluation, mitigations and released version.
Incorrect Answers
Red-team exercises show severe scalable misuse pathways without assigned mitigation owners. An auditor cannot identify the tested version or its risk-control evidence.
Admission staff intend to rely on model rankings to allocate limited places. An auditor cannot identify the tested version or its risk-control evidence.
Product shipments include no traceable conformity or responsible-operator documentation. An auditor cannot identify the tested version or its risk-control evidence.
Question 360
A retailer routes EU customer-support prompts to a service outside the approved region with no transfer review. Which additional observation would most strongly confirm the particular risk or control gap?
Explanation
Correct Answer
Network logs show personal prompts leave the expected territory through an undocumented inference endpoint.
Incorrect Answers
The vendor notice exists, but the lender has no record explaining the legal basis for its distinct profiling activity. Network logs show personal prompts leave the expected territory through an undocumented inference endpoint.
A transcript shows identifiable data disclosed to an unauthorized user and no breach triage record. Network logs show personal prompts leave the expected territory through an undocumented inference endpoint.
Reject decisions would become final without a person having authority to reconsider a case. Network logs show personal prompts leave the expected territory through an undocumented inference endpoint.
Artificial Intelligence Governance Professional (AIGP) Practice Tests
- IAPP AIGP Practice Test 34 - 10 Questions (Last Reviewed 19-Oct-2026)
- IAPP AIGP Practice Test 35 - 10 Questions (Last Reviewed 18-Oct-2026)
- IAPP AIGP Practice Test 36 - 10 Questions (Last Reviewed 17-Oct-2026)
- IAPP AIGP Practice Test 37 - 10 Questions (Last Reviewed 16-Oct-2026)
- IAPP AIGP Practice Test 38 - 10 Questions (Last Reviewed 15-Oct-2026)
