Free AAIA Quick Start: 10 Mixed Practice Questions

This diagnostic includes AI Governance and Risk, AI Operations, and AI Auditing Tools and Techniques. FlashGenius readiness results are guidance, not official ISACA passing scores.

Review all AAIA domains or begin the interactive quick-start test.

Quick-start questions with answers

Sample Question 1 — AI Governance and Risk

An enterprise AI policy requires material model changes to be independently reviewed before implementation. Management states that all changes followed the policy, but the change log shows several emergency updates to a fraud detection model during a high-volume attack. Which audit procedure is MOST appropriate?

  1. A. Select emergency changes and verify independent review, approval timing, and post-implementation ratification. (Correct answer)
  2. B. Interview the fraud operations manager to confirm the business need for emergency model updates.
  3. C. Inspect the AI policy to confirm whether emergency changes are permitted under defined conditions.
  4. D. Compare fraud loss trends before and after the emergency updates to assess business effectiveness.

Correct answer: A

Explanation: A is best because it directly tests operating effectiveness of the exception process for the highest-risk population: emergency changes. Verifying whether independent review occurred, whether approval timing complied with policy, and whether required ratification happened provides direct audit evidence. B is only inquiry and does not confirm the control operated. C evaluates policy design, not whether the control actually functioned for the emergency updates. D evaluates business outcome, which may be relevant operationally but does not evidence compliance with change governance requirements.

Sample Question 2 — AI Auditing Tools and Techniques

A telecommunications company uses AI to prioritize retention offers. Management provides a bias testing report with acceptable aggregate metrics, but several customer segments were excluded due to data limitations and the risk committee received only a summary. Which evidence BEST supports the conclusion that fairness risk was appropriately governed?

  1. A. Aggregate fairness results showing outcomes within management tolerance
  2. B. A summary presentation delivered to the risk committee
  3. C. Documented approval of exclusions and residual fairness risk (Correct answer)
  4. D. Management confirmation that excluded segments were immaterial

Correct answer: C

Explanation: C is the best answer because excluded segments create residual fairness risk that must be explicitly justified, reviewed, and accepted. Documented approval of the exclusions and the remaining fairness risk is the strongest evidence that governance addressed the incomplete testing scope. A is insufficient because favorable aggregate metrics can mask issues in omitted populations. B shows communication occurred, but not that assumptions and exclusions were reviewed and accepted. D is only management assertion and does not provide independent, supportable evidence of appropriate governance.

Sample Question 3 — AI Operations

A media company uses a third-party foundation model through an API to generate advertising content. The vendor may update the model without advance notice. Internal weekly tests are performed on sample prompts, and recent results have been acceptable. Which control deficiency is MOST significant?

  1. A. Vendor model changes are not subject to notification and impact assessment (Correct answer)
  2. B. Weekly prompt testing is not performed by an independent assurance function
  3. C. Sample prompt results are not retained for the full regulatory period
  4. D. Vendor service availability is not benchmarked against internal targets

Correct answer: A

Explanation: A is correct because the most significant gap is unmanaged third-party model change risk. If the vendor can change the underlying model without notice, outputs and compliance risk can change before the enterprise evaluates impact. Weekly testing helps, but it does not replace a control over change notification and assessment. B could strengthen assurance, but independence of weekly testing is less critical than not knowing when the model changes. C is an evidence-retention issue and secondary to unmanaged changes in model behavior. D concerns availability, which is not the primary AI risk described in the scenario.

Sample Question 4 — AI Auditing Tools and Techniques

An organization uses a vendor fairness dashboard for an AI recruiting platform. The dashboard reports no quarterly threshold breaches and is presented to the ethics committee. Internal audit notes that the dashboard monitors only resume screening results for broad gender categories, while the AI tool also ranks candidates and recommends interview slates. Complaints have increased in one business unit. Which control deficiency is MOST significant?

  1. A. Fairness monitoring does not cover key AI-influenced hiring stages (Correct answer)
  2. B. Quarterly ethics reports do not include complaint trend analysis
  3. C. Dashboard results are not independently recalculated before reporting
  4. D. Threshold settings are not refreshed after each reporting cycle

Correct answer: A

Explanation: A is best because the monitoring control is not designed to detect bias across material AI-influenced decision points. If ranking and interview recommendation stages are excluded, favorable results for resume screening alone cannot support a conclusion that fairness risk is being monitored effectively. B and C are plausible weaknesses that would reduce oversight quality, and D may become relevant over time, but those issues are less significant than a design gap that leaves major parts of the hiring workflow outside the control's coverage.

Sample Question 5 — AI Governance and Risk

A global consumer company uses an internal generative AI assistant for employee support. During a capacity issue, prompts containing customer information were routed to an external fallback AI service, triggering DLP alerts. IT operations closed the event within the service-level target after fixing the routing issue. The event was not reported to the AI governance committee or recorded in the AI risk register. Which recommendation is MOST appropriate?

  1. A. Revise incident criteria to require AI-specific escalation and reporting (Correct answer)
  2. B. Expand service-level reporting to include fallback routing outages
  3. C. Increase capacity testing evidence for the internal AI assistant
  4. D. Obtain management attestations for future routing exceptions

Correct answer: A

Explanation: A is correct because the scenario shows a design gap in incident classification and escalation: an AI-related event involving customer information and external routing was treated only as a routine IT incident. The control improvement needed is formal AI-specific escalation, governance reporting, and risk register tracking. B is not best because expanded SLA reporting improves operational visibility but does not ensure governance review. C addresses technical recurrence risk, not the core control weakness in incident governance. D is weaker because management attestations do not create a formal taxonomy, escalation trigger, or governance reporting process.

Sample Question 6 — AI Operations

During an audit of a healthcare diagnostic AI model, the auditor finds that model performance monitoring focuses on overall accuracy and latency, but there is no monitoring for data drift or concept drift. The model was trained on data from two years ago, and the hospital has recently changed diagnostic protocols. Which of the following is the MOST appropriate audit objective in this situation?

  1. A. To determine whether the AI model complies with GDPR requirements for data minimization and purpose limitation.
  2. B. To assess whether the AI model’s lifecycle management includes controls to detect and respond to data and concept drift that could impact clinical decision quality. (Correct answer)
  3. C. To verify that the AI model’s training data was obtained with appropriate patient consent and ethical approvals.
  4. D. To evaluate whether the AI model’s source code is protected against unauthorized modification in line with ISO 27001.

Correct answer: B

Explanation: The scenario highlights a change in clinical protocols and the absence of drift monitoring, which directly affects the reliability and safety of AI outputs in a healthcare context. In AI Operations, drift monitoring is critical to ensure models remain valid as environments and practices evolve. Option B is correct because it directly targets lifecycle management controls for detecting and responding to data and concept drift, which is the key operational risk in this scenario. This aligns with NIST AI RMF and ISO/IEC 42001 requirements for ongoing monitoring and performance management, especially for high-impact use cases like healthcare. Option A (GDPR data minimization and purpose limitation) is important but not the primary concern raised by the facts given. The issue is not about lawful basis or purpose, but about model degradation due to environmental changes. Option C (consent and ethics) is relevant to governance and compliance but does not address the operational risk that the model may now be misaligned with current clinical practice. Option D (source code protection) relates to security and integrity but does not address the risk that the model’s predictions may be clinically inappropriate due to drift. The PRIMARY audit objective should focus on lifecycle and monitoring controls for drift in this context.

Sample Question 7 — AI Auditing Tools and Techniques

A global insurer retrained an AI underwriting model after adding an external demographic data feed. The data lineage tool shows incomplete mappings for the new feed, while management states the feed was approved and model performance remains stable. What should the auditor evaluate FIRST?

  1. A. Whether recent model performance metrics remained within approved tolerance levels
  2. B. Whether the new feed is traceable to approved data source records (Correct answer)
  3. C. Whether management completed periodic attestations for external data usage
  4. D. Whether the underwriting team documented expected business benefits

Correct answer: B

Explanation: B is the best answer because the immediate audit issue is whether the newly added feed can be traced to an authorized source through reliable records. If traceability to approved source records cannot be demonstrated, the lineage control design is inadequate regardless of stable model performance. A is weaker because performance results do not prove data provenance. C is weaker because management attestations are less reliable than approved records and system-supported lineage evidence. D addresses business justification, not whether the source data feeding the model is controlled and auditable.

Sample Question 8 — AI Governance and Risk

A global employer uses an AI-assisted tool to screen candidates. The data science team performed fairness testing before launch, recruiters can override recommendations, and legal flagged the use case as sensitive. No evidence shows periodic independent challenge or documented acceptance of fairness thresholds. What is the PRIMARY audit concern?

  1. A. Fairness risk was not independently challenged or formally accepted (Correct answer)
  2. B. Recruiter override activity was not fully analyzed for all regions
  3. C. Prelaunch fairness testing was not repeated before each hiring cycle
  4. D. Legal sensitivity concerns were not tracked in the project issue log

Correct answer: A

Explanation: A is best because this is a sensitive employment use case, and the key governance question is whether fairness risk was independently challenged and formally approved for continued use. One-time testing and human override do not by themselves demonstrate sound governance. B and C may be valid monitoring improvements, and D is an issue-management weakness, but all are secondary to the absence of formal fairness governance and risk acceptance.

Sample Question 9 — AI Operations

A consumer bank uses a third-party LLM API to draft customer service responses. The provider supplies a control report and consistently meets uptime commitments. The bank retains limited request-response logs and has not tested manual fallback during provider outages. Which risk is MOST relevant?

  1. A. The bank may lack traceability and continuity despite vendor service assurances (Correct answer)
  2. B. The provider may change its underlying model without improving response quality
  3. C. The service desk may underreport low-severity errors in customer responses
  4. D. The agents may rely too heavily on drafted language during peak periods

Correct answer: A

Explanation: A is best because the stated control gaps are internal: the bank may be unable to reconstruct AI-assisted interactions or continue operations if the provider fails, even if vendor SLA metrics are strong. B is wrong because model-change quality risk is plausible but not the main deficiency described. C is wrong because error underreporting is less directly supported than the missing traceability and fallback evidence. D is wrong because agent reliance is relevant operationally, but it is not the primary third-party dependency risk in this scenario.

Sample Question 10 — AI Governance and Risk

A manufacturer routes AI initiatives through its standard IT risk assessment process. The process covers cybersecurity, availability, privacy, and vendor risk, but does not address AI-specific harms such as bias, explainability, model drift, or unintended use. Which control deficiency is MOST significant?

  1. A. The risk assessment taxonomy does not enable consistent evaluation of AI-specific risks (Correct answer)
  2. B. The IT risk assessment process has not been approved separately by the AI steering committee
  3. C. The assessment results are not summarized in a dashboard dedicated to AI initiatives
  4. D. The cybersecurity criteria are not weighted differently for internally developed AI models

Correct answer: A

Explanation: A is correct because if the risk taxonomy omits AI-specific harms, the organization cannot consistently identify, rate, and treat material AI risks even if conventional IT risks are assessed. That is the core design failure. B is a governance-formality issue and does not address the missing risk criteria. C is a reporting weakness that cannot compensate for an incomplete assessment method. D is too narrow because the problem is not cybersecurity weighting; it is the broader omission of AI-specific risk dimensions from the assessment itself.

AAIA Practice Test FAQs

What is the ISACA Advanced in AI Audit (AAIA) exam?

AAIA is an ISACA certification exam focused on auditing artificial intelligence systems. It covers AI governance and risk, AI operations, and AI auditing tools and techniques.

How many questions and how much time does the AAIA exam have?

The AAIA exam configuration is 90 questions in 150 minutes across three domains.

What score is required for AAIA?

ISACA uses a scaled score, with 450 on an 800-point scale listed as the passing score in the exam configuration. Practice-test percentages are not equivalent to an official ISACA scaled score.

How should I use AAIA practice questions?

Use mixed questions to find broad knowledge gaps, then use domain practice to review the concepts and explanations behind incorrect answers. FlashGenius readiness thresholds are study guidance, not official ISACA passing scores.

Which AAIA domain has the greatest weight?

AI Operations has the largest listed weight at 46%, followed by AI Governance and Risk at 33% and AI Auditing Tools and Techniques at 21%.

Explore AAIA Tests