IAPP AIGP Practice Test 33
Question 320
A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. An independent audit has requested current supporting evidence before expanded use. What is the most relevant fallback when the promised safeguard cannot be demonstrated?
- Purge excessive copies under the retention procedure and limit prompt access.
- Suspend disputed consequential outputs while correcting records and honoring review rights.
- Turn off unapproved emotion inference and delete unauthorized biometric derivatives where required.
- Disable cross-purpose analytics and remove unauthorized derived inferences where required.
Explanation
Correct Answer
C:Turn off unapproved emotion inference and delete unauthorized biometric derivatives where required. Correct. It limits further reliance or exposure while the risk remains unresolved. Biometric and sensitive-data uses need elevated legal scrutiny and cannot be excused by an analytics label.
Incorrect Answers
A:Purge excessive copies under the retention procedure and limit prompt access. Not selected. That response contains a different failure mode and leaves the stated risk unresolved. It belongs with Data minimization: Minimization applies across input, derived data, training sets, prompts and logs. The present case instead concerns Biometric and sensitive data.
B:Suspend disputed consequential outputs while correcting records and honoring review rights. Not selected. That response contains a different failure mode and leaves the stated risk unresolved. It belongs with Data subject rights and explainability: Privacy rights can affect model data, inferences and the decision pipeline, subject to applicable legal rules. The present case instead concerns Biometric and sensitive data.
D:Disable cross-purpose analytics and remove unauthorized derived inferences where required. Not selected. That response contains a different failure mode and leaves the stated risk unresolved. It belongs with Purpose limitation for inference: Derived AI inferences can be personal data and must be governed by their actual use. The present case instead concerns Biometric and sensitive data.
Question 321
A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review. What review step identifies the specific control gap rather than an unrelated AI risk?
- Assess whether biometric or other sensitive information is processed and whether specialized legal prohibitions or conditions apply.
- Identify jurisdictions, remote access, subprocessor roles and the transfer mechanism that applies.
- Map each retained field to a specific necessity and evaluate less intrusive alternatives.
- Document an appropriate lawful basis and compatibility assessment, then limit training to authorized data.
Explanation
Correct Answer
A:Assess whether biometric or other sensitive information is processed and whether specialized legal prohibitions or conditions apply. Correct. It establishes the relevant factual risk boundary before implementation. Biometric and sensitive-data uses need elevated legal scrutiny and cannot be excused by an analytics label.
Incorrect Answers
B:Identify jurisdictions, remote access, subprocessor roles and the transfer mechanism that applies. Not selected. That check primarily diagnoses a different governance issue. It belongs with Cross-border processing: AI data flows can create cross-border transfers through prompts, storage, support access and subprocessors. The present case instead concerns Biometric and sensitive data.
C:Map each retained field to a specific necessity and evaluate less intrusive alternatives. Not selected. That check primarily diagnoses a different governance issue. It belongs with Data minimization: Minimization applies across input, derived data, training sets, prompts and logs. The present case instead concerns Biometric and sensitive data.
D:Document an appropriate lawful basis and compatibility assessment, then limit training to authorized data. Not selected. This action relates to Lawful basis in model training. AI training does not erase lawful basis and purpose-limitation obligations attaching to personal data. For the diagnose decision, it does not resolve the specifically described issue of Biometric and sensitive data in A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review.
Question 322
A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review. Which safeguard most directly treats the described failure mode?
- Extend rights-handling workflows to correctable model inputs and legally applicable automated-decision safeguards.
- Review transfer safeguards and update contracts, notices and technical routing restrictions before the change.
- Perform a strict necessity and legal-risk assessment, and reject the function unless genuinely permitted with safeguards.
- Determine the processing purpose, original disclosures, lawful basis and whether the new training use is compatible with applicable privacy law.
Explanation
Correct Answer
C:Perform a strict necessity and legal-risk assessment, and reject the function unless genuinely permitted with safeguards. Correct. It directly reduces or governs the specific documented risk. Biometric and sensitive-data uses need elevated legal scrutiny and cannot be excused by an analytics label.
Incorrect Answers
A:Extend rights-handling workflows to correctable model inputs and legally applicable automated-decision safeguards. Not selected. That action treats another control objective, not the specific gap described. It belongs with Data subject rights and explainability: Privacy rights can affect model data, inferences and the decision pipeline, subject to applicable legal rules. The present case instead concerns Biometric and sensitive data.
B:Review transfer safeguards and update contracts, notices and technical routing restrictions before the change. Not selected. That action treats another control objective, not the specific gap described. It belongs with Cross-border processing: AI data flows can create cross-border transfers through prompts, storage, support access and subprocessors. The present case instead concerns Biometric and sensitive data.
D:Determine the processing purpose, original disclosures, lawful basis and whether the new training use is compatible with applicable privacy law. Not selected. This action relates to Lawful basis in model training. AI training does not erase lawful basis and purpose-limitation obligations attaching to personal data. For the control decision, it does not resolve the specifically described issue of Biometric and sensitive data in A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review.
Question 323
A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review. Which verification task should support the next approval decision?
- Test retention, access and purpose restrictions and review applicability of special-category requirements.
- Review sample prompt payloads and logs to confirm only required elements remain.
- Exercise a test request and verify the system propagates corrections or documents lawful limits.
- Document an appropriate lawful basis and compatibility assessment, then limit training to authorized data.
Explanation
Correct Answer
A:Test retention, access and purpose restrictions and review applicability of special-category requirements. Correct. It yields measurable evidence that the case-specific safeguard works. Biometric and sensitive-data uses need elevated legal scrutiny and cannot be excused by an analytics label.
Incorrect Answers
B:Review sample prompt payloads and logs to confirm only required elements remain. Not selected. That test would not demonstrate this particular risk has been addressed. It belongs with Data minimization: Minimization applies across input, derived data, training sets, prompts and logs. The present case instead concerns Biometric and sensitive data.
C:Exercise a test request and verify the system propagates corrections or documents lawful limits. Not selected. That test would not demonstrate this particular risk has been addressed. It belongs with Data subject rights and explainability: Privacy rights can affect model data, inferences and the decision pipeline, subject to applicable legal rules. The present case instead concerns Biometric and sensitive data.
D:Document an appropriate lawful basis and compatibility assessment, then limit training to authorized data. Not selected. This action relates to Lawful basis in model training. AI training does not erase lawful basis and purpose-limitation obligations attaching to personal data. For the validate decision, it does not resolve the specifically described issue of Biometric and sensitive data in A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review.
Question 324
A stadium wants to classify attendee emotions from facial footage, treating the feature as ordinary anonymous analytics. A supplier or jurisdictional change requires a fresh legal-control review. Select TWO complementary actions: first assess the stated concern and then implement its targeted control.
- Perform a strict necessity and legal-risk assessment, and reject the function unless genuinely permitted with safeguards.
- Assess whether biometric or other sensitive information is processed and whether specialized legal prohibitions or conditions apply.
- Review transfer safeguards and update contracts, notices and technical routing restrictions before the change.
- Compare the new inferential use with the original collection notice, purpose, expectations and applicable law.
Explanation
Correct Answers
A:Perform a strict necessity and legal-risk assessment, and reject the function unless genuinely permitted with safeguards. Correct. As the second, implementation half of the requested pair, this directly applies the needed safeguard. Biometric and sensitive-data uses need elevated legal scrutiny and cannot be excused by an analytics label.
B:Assess whether biometric or other sensitive information is processed and whether specialized legal prohibitions or conditions apply. Correct. As the first, evidence-gathering half of the requested pair, this establishes the specific risk boundary. Biometric and sensitive-data uses need elevated legal scrutiny and cannot be excused by an analytics label.
Incorrect Answers
C:Review transfer safeguards and update contracts, notices and technical routing restrictions before the change. Not selected. It addresses a different governance mechanism, not the required assessment-and-control pair. It belongs with Cross-border processing: AI data flows can create cross-border transfers through prompts, storage, support access and subprocessors. The present case instead concerns Biometric and sensitive data.
D:Compare the new inferential use with the original collection notice, purpose, expectations and applicable law. Not selected. It addresses a different governance mechanism, not the required assessment-and-control pair. It belongs with Purpose limitation for inference: Derived AI inferences can be personal data and must be governed by their actual use. The present case instead concerns Biometric and sensitive data.
Question 325
A model developer scrapes paid textbooks and claims any accessible website content is free for AI training. Legal and AI governance staff must agree on the defensible compliance response. What review step identifies the specific control gap rather than an unrelated AI risk?
- Determine source licenses, access conditions, copyright exceptions and jurisdiction-specific permissions.
- Investigate disparate error rates, feature proxies and applicable employment nondiscrimination requirements.
- Examine whether the feature yields disproportionate outcomes under applicable lending rules.
- Compare marketing claims with measured capabilities, limitations and consumer-protection standards.
Explanation
Correct Answer
A:Determine source licenses, access conditions, copyright exceptions and jurisdiction-specific permissions. Correct. It establishes the relevant factual risk boundary before implementation. Public accessibility does not automatically grant permission to reproduce or train on protected material.
Incorrect Answers
B:Investigate disparate error rates, feature proxies and applicable employment nondiscrimination requirements. Not selected. That check primarily diagnoses a different governance issue. It belongs with Employment discrimination: Equal treatment obligations apply to algorithmic employment decisions and proxy discrimination. The present case instead concerns Training-data copyright.
C:Examine whether the feature yields disproportionate outcomes under applicable lending rules. Not selected. That check primarily diagnoses a different governance issue. It belongs with Credit and lending fairness: AI credit decisions remain subject to lending fairness, transparency and contestability obligations. The present case instead concerns Training-data copyright.
D:Compare marketing claims with measured capabilities, limitations and consumer-protection standards. Not selected. That check primarily diagnoses a different governance issue. It belongs with Consumer deception: Consumer protection law can apply to AI representations even without an AI-specific statute. The present case instead concerns Training-data copyright.
Question 326
A model developer scrapes paid textbooks and claims any accessible website content is free for AI training. Legal and AI governance staff must agree on the defensible compliance response. Which safeguard most directly treats the described failure mode?
- Require human clearance and vendor-license review for commercial deliverables before release.
- Add job-relevant validation, bias mitigation, documented accommodations and accountable human review.
- Require validated fail-safe design, representative safety testing and defect-management controls.
- Obtain rights or use lawfully eligible content and retain a traceable licensing inventory.
Explanation
Correct Answer
D:Obtain rights or use lawfully eligible content and retain a traceable licensing inventory. Correct. It directly reduces or governs the specific documented risk. Public accessibility does not automatically grant permission to reproduce or train on protected material.
Incorrect Answers
A:Require human clearance and vendor-license review for commercial deliverables before release. Not selected. That action treats another control objective, not the specific gap described. It belongs with Generated-output IP: AI output availability does not guarantee ownership, exclusivity or freedom from infringement. The present case instead concerns Training-data copyright.
B:Add job-relevant validation, bias mitigation, documented accommodations and accountable human review. Not selected. That action treats another control objective, not the specific gap described. It belongs with Employment discrimination: Equal treatment obligations apply to algorithmic employment decisions and proxy discrimination. The present case instead concerns Training-data copyright.
C:Require validated fail-safe design, representative safety testing and defect-management controls. Not selected. That action treats another control objective, not the specific gap described. It belongs with Product safety liability: AI systems inside products can raise ordinary safety and product-liability risks. The present case instead concerns Training-data copyright.
Question 327
A model developer scrapes paid textbooks and claims any accessible website content is free for AI training. Legal and AI governance staff must agree on the defensible compliance response. Which verification task should support the next approval decision?
- Test hazardous edge cases and record whether safe-stop performance meets the accepted criterion.
- Reevaluate relevant subgroup outcomes and lawful reason notices after the change.
- Run targeted similarity checks against known protected brands and retain reviewer signoff.
- Sample training records to verify each restricted source has a documented permitted basis.
Explanation
Correct Answer
D:Sample training records to verify each restricted source has a documented permitted basis. Correct. It yields measurable evidence that the case-specific safeguard works. Public accessibility does not automatically grant permission to reproduce or train on protected material.
Incorrect Answers
A:Test hazardous edge cases and record whether safe-stop performance meets the accepted criterion. Not selected. That test would not demonstrate this particular risk has been addressed. It belongs with Product safety liability: AI systems inside products can raise ordinary safety and product-liability risks. The present case instead concerns Training-data copyright.
B:Reevaluate relevant subgroup outcomes and lawful reason notices after the change. Not selected. That test would not demonstrate this particular risk has been addressed. It belongs with Credit and lending fairness: AI credit decisions remain subject to lending fairness, transparency and contestability obligations. The present case instead concerns Training-data copyright.
C:Run targeted similarity checks against known protected brands and retain reviewer signoff. Not selected. That test would not demonstrate this particular risk has been addressed. It belongs with Generated-output IP: AI output availability does not guarantee ownership, exclusivity or freedom from infringement. The present case instead concerns Training-data copyright.
Question 328
A model developer scrapes paid textbooks and claims any accessible website content is free for AI training. Legal and AI governance staff must agree on the defensible compliance response. If the control fails, which immediate response most appropriately limits the documented risk?
- Remove unauthorized sources and seek legal remediation before model release.
- Withdraw suspect assets and stop distribution while infringement risks are assessed.
- Suspend automated adverse screening where evidence suggests discriminatory impact.
- Disable the unsafe autonomy mode and initiate product incident procedures if safety thresholds fail.
Explanation
Correct Answer
A:Remove unauthorized sources and seek legal remediation before model release. Correct. It limits further reliance or exposure while the risk remains unresolved. Public accessibility does not automatically grant permission to reproduce or train on protected material.
Incorrect Answers
B:Withdraw suspect assets and stop distribution while infringement risks are assessed. Not selected. That response contains a different failure mode and leaves the stated risk unresolved. It belongs with Generated-output IP: AI output availability does not guarantee ownership, exclusivity or freedom from infringement. The present case instead concerns Training-data copyright.
C:Suspend automated adverse screening where evidence suggests discriminatory impact. Not selected. That response contains a different failure mode and leaves the stated risk unresolved. It belongs with Employment discrimination: Equal treatment obligations apply to algorithmic employment decisions and proxy discrimination. The present case instead concerns Training-data copyright.
D:Disable the unsafe autonomy mode and initiate product incident procedures if safety thresholds fail. Not selected. That response contains a different failure mode and leaves the stated risk unresolved. It belongs with Product safety liability: AI systems inside products can raise ordinary safety and product-liability risks. The present case instead concerns Training-data copyright.
Question 329
A model developer scrapes paid textbooks and claims any accessible website content is free for AI training. An independent audit has requested current supporting evidence before expanded use. Which initial evidence check most directly addresses the governance concern?
- Investigate disparate error rates, feature proxies and applicable employment nondiscrimination requirements.
- Review potential copyright, trademark and contractual conflicts in the intended commercial use.
- Determine source licenses, access conditions, copyright exceptions and jurisdiction-specific permissions.
- Examine whether the feature yields disproportionate outcomes under applicable lending rules.
Explanation
Correct Answer
C:Determine source licenses, access conditions, copyright exceptions and jurisdiction-specific permissions. Correct. It establishes the relevant factual risk boundary before implementation. Public accessibility does not automatically grant permission to reproduce or train on protected material.
Incorrect Answers
A:Investigate disparate error rates, feature proxies and applicable employment nondiscrimination requirements. Not selected. That check primarily diagnoses a different governance issue. It belongs with Employment discrimination: Equal treatment obligations apply to algorithmic employment decisions and proxy discrimination. The present case instead concerns Training-data copyright.
B:Review potential copyright, trademark and contractual conflicts in the intended commercial use. Not selected. That check primarily diagnoses a different governance issue. It belongs with Generated-output IP: AI output availability does not guarantee ownership, exclusivity or freedom from infringement. The present case instead concerns Training-data copyright.
D:Examine whether the feature yields disproportionate outcomes under applicable lending rules. Not selected. That check primarily diagnoses a different governance issue. It belongs with Credit and lending fairness: AI credit decisions remain subject to lending fairness, transparency and contestability obligations. The present case instead concerns Training-data copyright.
Question 330
A model developer scrapes paid textbooks and claims any accessible website content is free for AI training. An independent audit has requested current supporting evidence before expanded use. Which specific implementation best addresses the identified gap?
- Remove unjustified proxy features and establish defensible credit-relevance and adverse-action review controls.
- Add job-relevant validation, bias mitigation, documented accommodations and accountable human review.
- Replace absolute claims with substantiated representations and conspicuous material limitations.
- Obtain rights or use lawfully eligible content and retain a traceable licensing inventory.
Explanation
Correct Answer
D:Obtain rights or use lawfully eligible content and retain a traceable licensing inventory. Correct. It directly reduces or governs the specific documented risk. Public accessibility does not automatically grant permission to reproduce or train on protected material.
Incorrect Answers
A:Remove unjustified proxy features and establish defensible credit-relevance and adverse-action review controls. Not selected. That action treats another control objective, not the specific gap described. It belongs with Credit and lending fairness: AI credit decisions remain subject to lending fairness, transparency and contestability obligations. The present case instead concerns Training-data copyright.
B:Add job-relevant validation, bias mitigation, documented accommodations and accountable human review. Not selected. That action treats another control objective, not the specific gap described. It belongs with Employment discrimination: Equal treatment obligations apply to algorithmic employment decisions and proxy discrimination. The present case instead concerns Training-data copyright.
C:Replace absolute claims with substantiated representations and conspicuous material limitations. Not selected. That action treats another control objective, not the specific gap described. It belongs with Consumer deception: Consumer protection law can apply to AI representations even without an AI-specific statute. The present case instead concerns Training-data copyright.