ISACA AAISM Exam Questions
ISACA Advanced in AI Security Management (Page 4 )

Updated On: 3-Jun-2026

Which of the following technologies can be used to manage deepfake risk?

  1. Systematic data tagging
  2. Multi-factor authentication (MFA)
  3. Blockchain
  4. Adaptive authentication

Answer(s): C

Explanation:

The AAISM study material highlights blockchain as a control mechanism for managing deepfake risk because it provides immutable verification of digital media provenance. By anchoring original data signatures on a blockchain, organizations can verify authenticity and detect tampered or synthetic content. Data tagging helps organize but does not guarantee authenticity. MFA and adaptive authentication strengthen identity security but do not address content manipulation risks.

Blockchain's immutability and traceability make it the recognized technology for mitigating deepfake challenges.


Reference:

AAISM Study Guide ­ AI Technologies and Controls (Emerging Controls for Content Authenticity)

ISACA AI Governance Guidance ­ Blockchain for Data Integrity and Deepfake Mitigation



Which of the following would BEST help mitigate vulnerabilities associated with hidden triggers in generative AI models?

  1. Regularly retraining the model using a diverse data set
  2. Applying differential privacy and masking sensitive patterns in the training data
  3. Incorporating adversarial training to expose and neutralize potential triggers
  4. Monitoring model outputs and suspicious patterns to detect trigger activations

Answer(s): C

Explanation:

Hidden triggers are adversarial backdoors planted in AI models, activated only by specific inputs. The AAISM materials specify that the best mitigation is to use adversarial training, which deliberately exposes the model to potential trigger inputs during training so it can learn to neutralize or resist them. Retraining with diverse data reduces bias but does not address hidden triggers. Differential privacy is focused on privacy preservation, not adversarial resilience. Monitoring outputs can help with detection but is reactive rather than preventative. The proactive solution highlighted in the study guide is adversarial training.


Reference:

AAISM Exam Content Outline ­ AI Risk Management (Backdoors and Hidden Triggers)

AI Security Management Study Guide ­ Adversarial Training as a Mitigation Control



An organization plans to apply an AI system to its business, but developers find it difficult to predict system results due to lack of visibility to the inner workings of the AI model.
Which of the following is the GREATEST challenge associated with this situation?

  1. Gaining the trust of end users through explainability and transparency
  2. Assigning a risk owner who is responsible for system uptime and performance
  3. Determining average turnaround time for AI transaction completion
  4. Continuing operations to meet expected AI security requirements

Answer(s): A

Explanation:

AAISM materials identify explainability and transparency as the greatest challenge when models operate as "black boxes" where inner logic is opaque. Inability to interpret how results are produced undermines the trust of business users, customers, regulators, and auditors. Explainability is emphasized as a critical governance requirement, because without it, ethical validation, accountability, and regulatory compliance are at risk. Assigning risk owners or measuring transaction times are operational concerns, but they do not address the core trust deficit caused by lack of visibility. The greatest challenge in this situation is therefore the loss of end-user trust due to insufficient explainability.


Reference:

AAISM Study Guide ­ AI Governance and Program Management (Transparency and Explainability)

ISACA AI Security Management ­ Ethical and Trust Considerations



Embedding unique identifiers into AI models would BEST help with:

  1. Preventing unauthorized access
  2. Tracking ownership
  3. Eliminating AI system biases
  4. Detecting adversarial attacks

Answer(s): B

Explanation:

The AAISM framework explains that embedding unique identifiers--such as digital watermarks or model fingerprints--enables organizations to trace and verify model provenance. This technique is used for tracking ownership and intellectual property rights over models, particularly when sharing, licensing, or distributing AI systems.
While identifiers may support certain security functions, their primary control objective is ownership verification, not preventing access, bias removal, or adversarial detection. The correct alignment with AAISM controls is tracking ownership.


Reference:

AAISM Exam Content Outline ­ AI Technologies and Controls (Model Provenance and Watermarking)

AI Security Management Study Guide ­ Ownership and Accountability of Models



Which of the following BEST describes the role of risk documentation in an AI governance program?

  1. Providing a record of past AI-related incidents for audits
  2. Outlining the acceptable levels of risk for AI-related initiatives
  3. Offering detailed analyses of technical risk and vulnerabilities
  4. Demonstrating governance, risk, and compliance (GRC) for external stakeholders

Answer(s): B

Explanation:

In AAISM governance guidance, risk documentation is described as the structured record that defines the organization's risk appetite and tolerance levels for AI initiatives. By outlining acceptable levels of risk, documentation ensures decision-makers can approve, monitor, and adjust AI projects within defined boundaries.
While it may also serve audit functions, technical analysis, or communication to stakeholders, its primary role is to formalize risk acceptance thresholds and integrate them into governance and decision-making. This aligns directly with the governance requirement to align AI adoption with organizational risk appetite.


Reference:

AAISM Study Guide ­ AI Governance and Program Management (Risk Documentation and Appetite)

ISACA AI Security Management ­ Governance, Risk and Compliance Integration



In the context of generative AI, which of the following would be the MOST likely goal of penetration testing during a red-teaming exercise?

  1. Generate outputs that are unexpected using adversarial inputs
  2. Stress test the model's decision-making process
  3. Degrade the model's performance for existing use cases
  4. Replace the model's outputs with entirely random content

Answer(s): A

Explanation:

AAISM's risk management content describes red-teaming in generative AI as focused on deliberately crafting adversarial prompts to test whether the model produces unexpected or undesired outputs that violate safety, integrity, or compliance standards. The goal is not to stress system performance or randomly disrupt outputs, but rather to uncover vulnerabilities in how the model responds to manipulative inputs. This allows organizations to improve resilience against prompt injection, jailbreaking, or harmful content generation. The correct answer is therefore generate outputs that are unexpected using adversarial inputs.


Reference:

AAISM Exam Content Outline ­ AI Risk Management (Red-Team Testing and Adversarial Exercises)

AI Security Management Study Guide ­ Penetration Testing in Generative AI Contexts



An organization needs large data sets to perform application testing.
Which of the following would BEST fulfill this need?

  1. Reviewing AI model cards
  2. Incorporating data from search content
  3. Using open-source data repositories
  4. Performing AI data augmentation

Answer(s): C

Explanation:

According to AAISM study guidance, the most direct and effective way to obtain large volumes of diverse data for application testing is through open-source data repositories. These repositories provide freely available, well-documented, and often standardized data that supports testing and benchmarking in a compliant manner. Model cards document AI behavior but do not provide data.

Incorporating search content may introduce legal, privacy, and quality risks. Data augmentation is useful for expanding existing sets but does not provide the breadth or size required when starting with insufficient data. The recommended best practice for sourcing large testing datasets is therefore the use of open-source repositories.


Reference:

AAISM Study Guide ­ AI Technologies and Controls (Data Sources and Testing Practices)

ISACA AI Security Management ­ Data Governance and Compliance in AI Testing



When integrating AI for innovation, which of the following can BEST help an organization manage security risk?

  1. Re-evaluating the risk appetite
  2. Seeking third-party advice
  3. Evaluating compliance requirements
  4. Adopting a phased approach

Answer(s): D

Explanation:

AAISM emphasizes that when introducing innovative AI systems, organizations reduce security and compliance risk by following a phased adoption approach. This allows incremental deployment, controlled testing, and gradual scaling while monitoring risks in real time. Re-evaluating risk appetite and evaluating compliance are important governance steps but do not directly mitigate risks during implementation. Seeking third-party advice can add expertise but does not provide the structured control that phased integration offers. The most effective risk management approach for AI innovation is to adopt a phased rollout strategy.


Reference:

AAISM Exam Content Outline ­ AI Risk Management (Innovation and Risk Control)

AI Security Management Study Guide ­ Phased Implementation Strategies



Viewing page 4 of 33
Viewing questions 25 - 32 out of 255 questions


AAISM Exam Discussions & Posts (Share your experience with others)

AI Tutor AI Tutor 👋 I’m here to help!