Which of the following should be the MOST significant concern related to datasets used for training AI models?
Answer(s): D
Lack of documentation on data sources prevents verification of data origin, quality, and appropriateness, increasing the risk of bias, noncompliance, and untraceable errors, which undermines governance, auditability, and model reliability.
Which of the following AI governance program objectives BEST supports transparency?
Making AI algorithm rationales accessible ensures stakeholders can understand how decisions are made, directly supporting transparency by providing insight into model logic, assumptions, and outputs for audit, regulatory, and ethical purposes.
An organization intends to implement an AI system to enhance customer service efficiency.Which of the following actions is MOST appropriate to ensure the AI initiative aligns with organizational goals and governance requirements?
Defining a clear AI business use case establishes the purpose, scope, processes, and stakeholder involvement, ensuring that the AI initiative supports organizational objectives, aligns with governance requirements, and provides measurable value while mitigating operational and compliance risks.
A financial institution has an AI-based credit screening system that is designed to exclude gender, race, and sexual orientation from collected customer information.To ensure the effectiveness of this control, which of the following should be an IS auditor’s PRIMARY focus?
Answer(s): B
Even if sensitive attributes are excluded, correlated features can inadvertently introduce bias. The auditor should focus on whether allowable attributes could serve as proxies for excluded information, ensuring that the AI system does not indirectly discriminate and that controls effectively mitigate bias.
Which of the following is the MOST appropriate role for an IS auditor involved in the pre-implementation review of a new AI model?
Answer(s): C
The auditor’s role is to provide independent assurance that industry best practices, governance, risk management, and compliance requirements are followed in AI model development, without becoming directly involved in technical design or implementation decisions.
Post your Comments and Discuss ISACA AAIA exam prep with other Community members:
Our website is free, but we have to fight against AI bots and content theft. We're sorry for the inconvenience caused by these security measures. You can access the rest of the AAIA content, but please register or login to continue.