Isaca AAISM Online Practice Questions and Exam Preparation

Isaca AAISM Online Questions & Answers

• Question 101:

  Which of the following is the MOST important course of action when implementing continuous monitoring and reporting for AI-based systems?

  A. Establish an automated alert system for threshold breaches in risk metrics
  B. Develop standardized risk reporting templates for different stakeholder groups
  C. Implement real-time monitoring of key risk indicators (KRIs) for AI systems
  D. Implement a risk dashboard for visualizing and tracking AI-related risk over time
  C. Implement real-time monitoring of key risk indicators (KRIs) for AI systems

  ---

  The AAISM governance framework specifies that the foundation of continuous monitoring is real-time tracking of key risk indicators. This ensures immediate detection of deviations, model drift, and operational anomalies. Automated alerts, dashboards, and reporting templates all support monitoring, but they rely on the presence of accurate, real-time KRI measurement as their source. Without live monitoring, the other controls are reactive rather than proactive. The most important course of action in

  establishing effective continuous monitoring is therefore real-time KRI tracking.

  References:

  AAISM Study Guide - AI Governance and Program Management (Continuous Monitoring and Assurance)

  ISACA AI Risk Guidance - Monitoring Key Risk Indicators

• Question 102:

  An organization is adopting an agentic AI solution from an external vendor to support its internal IT operations. To evaluate the security posture of this system, which of the following provides the MOST reliable and independently verifiable evidence of implemented security controls?

  A. Internal red team testing reports
  B. Industry benchmarking peer review
  C. General AI security whitepapers
  D. Third-party audit reports
  D. Third-party audit reports

  ---

  Third-party audit reports provide independent assurance that the vendor's stated controls are designed and operating effectively against recognized criteria. Such attestations (e.g., audit/assurance frameworks) are traceable, repeatable, and verifiable, and they support supply-chain risk reviews and contractual assurance. Internal red-team reports are not independent, industry "peer reviews" are not control attestations, and whitepapers are marketing/educational materials without evidence of control operation.

  References: AI Security Management. (AAISM) Body of Knowledge: Third-Party and Supply-Chain Assurance; Independent Assessment and Audit Evidence; Vendor Risk Management. AAISM Study Guide: Assurance Artifacts for AI Services; Evaluating External AI Providers.

• Question 103:

  Cybersecurity teams should FIRST be embedded in the:

  A. Model testing phase
  B. Model deployment phase
  C. Model training phase
  D. Model design phase
  D. Model design phase

  ---

  AAISM stresses that security must be embedded from the earliest phase-design, ensuring: . threat modeling . secure architecture . data protection requirements . system boundaries . security-by-design principles Introducing cybersecurity later increases unmitigated vulnerabilities. Testing (A), training (C), and deployment (B) occur after foundational security decisions have already been

  made.

  References: AAISM Study Guide - Secure AI Lifecycle; Early Security Involvement.

• Question 104:

  Which of the following is the GREATEST concern when a vendor enables generative AI features for an organization's critical system?

  A. Access to the model
  B. Proposed regulatory enhancements
  C. Security monitoring and alerting
  D. Bias and ethical practices
  A. Access to the model

  ---

  When enabling genAI capabilities in a critical system, AAISM prioritizes controlling access to the model and its interfaces (prompt surfaces, context windows, tools/functions, and connected data) because exposure expands the attack surface for prompt injection, data exfiltration, jailbreaks, and misuse. Monitoring (C) is necessary but detective; ethics and bias (D) are vital but secondary to immediate safety and security of a mission-critical environment; proposed regulations (B) are not an immediate operational risk.

  References: AAISM Body of Knowledge: GenAI Security-Access Governance, Interface Hardening, and Prompt Surface Controls; AAISM Study Guide: Critical System Safeguards-Least Privilege, Guardrails, and Abuse Prevention.

• Question 105:

  An attack has occurred on an AI system that has been in use for two years. Which of the following would BEST mitigate the impact of the attack?

  A. Monitoring AI systems for suspicious activities
  B. Updating deployed training data with new adversarial data
  C. Replacing the AI model with a new model that hides confidence levels
  D. Implementing strict access controls to the model's architecture
  B. Updating deployed training data with new adversarial data

  ---

  When an AI system experiences an attack after being in production for an extended period, the most effective mitigation strategy is to update the deployed training data with new adversarial data. This process strengthens the model's resilience by retraining it to recognize and resist attack vectors that were previously unknown or unaccounted for. According to the AI Security Management (AAISM) framework, risk mitigation for AI systems must address model robustness through adversarial retraining, data

  quality improvement, and model lifecycle hardening rather than relying solely on reactive measures.

  Why Option B is Correct:

  Incorporating adversarial examples into the training set enhances the system's ability to correctly classify and withstand malicious inputs.

  This approach directly mitigates the vulnerability exploited in the attack and supports a proactive, continuous risk management cycle.

  Why Other Options Are Incorrect:

  Option A: Monitoring helps detect suspicious activity but does not resolve the underlying vulnerability.

  Option C: Concealing confidence scores may reduce model transparency but does not address the attack mechanism or its root cause.

  Option D: Implementing access controls protects the model's architecture but does not improve model robustness against input manipulation attacks.

  Exact Extract from Official AAISM Study Guide:

  "AI risk management requires continuous improvement following incidents. After an adversarial or data poisoning event, the preferred risk treatment involves retraining the model using adversarial data and updated datasets to enhance robustness. This ensures the AI model adapts to evolving threat landscapes rather than merely restricting access or obscuring outputs."

  References:

  AI Security Management (AAISM) Body of Knowledge: AI Risk Treatment and Mitigation Strategies, Adversarial Robustness and Resilience Engineering.

  AI Security Management Study Guide: Model Lifecycle Security, Continuous Risk Treatment through Adversarial Retraining.

  ISO/IEC 23894:2023, Clause 8.3.2 - Risk treatment through robustness improvement and adversarial data inclusion.

• Question 106:

  Which of the following BEST enables an organization to maintain visibility to its AI usage?

  A. Ensuring the board approves the policies and standards that define corporate AI strategy
  B. Maintaining a monthly dashboard that captures all AI vendors
  C. Maintaining a comprehensive inventory of AI systems and business units that leverage them
  D. Measuring the impact of AI implementation using key performance indicators (KPIs)
  C. Maintaining a comprehensive inventory of AI systems and business units that leverage them

  ---

  The AAISM framework stresses that the most effective way to maintain oversight of organizational AI usage is by maintaining a comprehensive inventory of all AI systems and the business units using them. Such an inventory provides a centralized, transparent record of where AI is deployed, ensuring accountability, monitoring, and compliance. While board approval, dashboards, and KPIs are important governance tools, they do not provide holistic visibility across the enterprise. The inventory ensures

  traceability and governance alignment, making it the best method to maintain visibility of AI usage.

  References:

  AAISM Study Guide - AI Governance and Program Management (AI Inventories)

  ISACA AI Security Management - Centralized Oversight of AI Assets

• Question 107:

  Implementing which of the following would MOST effectively address bias in generative AI models?

  A. Data augmentation
  B. Data minimization
  C. Adversarial training
  D. Fairness constraints
  D. Fairness constraints

  ---

  AAISM identifies fairness constraints (e.g., constrained optimization, debiasing objectives, conditional generation controls, and post-processing calibrations) as the most direct, measurable method to mitigate disparate outcomes in generative systems. While data augmentation can help with coverage, and adversarial training improves robustness, fairness constraints explicitly target distributional fairness and outcome equity in generated content, aligning with governance and compliance goals.

  References: AI Security Management (AAISM) Body of Knowledge - Fairness and Bias Management in Generative AI; Metrics, Constraints, and Remediation. AAISM Study Guide - Fairness Objectives, Post-hoc Debiasing, and Evaluation Protocols.

• Question 108:

  Who is responsible for implementing recommendations in a final report after an external AI compliance audit?

  A. System architects
  B. Internal auditors
  C. End users
  D. Model owners
  D. Model owners

  ---

  AAISM clarifies that model owners hold responsibility for ensuring corrective actions are implemented after AI audits. They are accountable for: model behavior compliance gaps security improvements governance alignment Internal auditors (B) perform assessments but do not implement changes. System architects (A) support technical fixes but do not own compliance. End users (C) are not responsible for audit remediation.

  References: AAISM Study Guide - Roles and Responsibilities; AI Ownership and Accountability.

• Question 109:

  Which of the following security framework elements BEST helps to safeguard the integrity of outputs generated by AI algorithms?

  A. Risk exposure due to bias in AI outputs is kept within an acceptable range
  B. Ethical standards are incorporated into security awareness programs
  C. Management is prepared to disclose AI system architecture to stakeholders
  D. Responsibility is defined for legal actions related to AI regulatory requirements
  A. Risk exposure due to bias in AI outputs is kept within an acceptable range

  ---

  According to AAISM technical controls, the element of security frameworks that directly safeguards output integrity is ensuring that bias-related risks are maintained within acceptable ranges. This ensures that AI results remain consistent, fair, and reliable, preserving trust in outputs. Ethical awareness, architectural disclosure, and legal responsibilities are governance practices but do not directly secure output integrity. Output integrity is primarily protected through bias management and ongoing evaluation of

  fairness and accuracy.

  References:

  AAISM Exam Content Outline - AI Technologies and Controls (Integrity of AI Outputs)

  AI Security Management Study Guide - Bias and Output Integrity Controls

• Question 110:

  Which of the following BEST ensures the integrity of data sets used to train AI models?

  A. Collection and retention of only necessary data sets
  B. Tracking and verification of data sets via cryptographic controls
  C. Appropriate storage of data sets according to documented classification processes
  D. Clear documentation of data sources, types used, and processing steps
  B. Tracking and verification of data sets via cryptographic controls

  ---

  AAISM defines cryptographic tracking and verification as the best control for ensuring the integrity of training data. By applying hashing and verification methods, organizations can confirm that datasets remain unaltered and authentic throughout collection, storage, and processing. Collecting only necessary data, proper storage, or clear documentation all support governance and compliance, but they do not guarantee that the data has not been tampered with. Integrity is specifically ensured by cryptographic

  verification techniques.

  References:

  AAISM Exam Content Outline - AI Risk Management (Data Integrity and Protection)

  AI Security Management Study Guide - Cryptographic Controls for Dataset Integrity