This is not a replacement for the official study guide. Please read the official study guide first — use this exam prep only to reinforce or practice your understanding.
Free AAISM exam practice questions with answers and explanations, organized by domain and part.
Domain 2: AI Risk Management
Q341 Data from an original source is moved into an AI solution. What privacy control is MOST important?
- A. Ensure data protection principles from the original location persist in the AI solution
- B. Share the data with all partners to improve model accuracy
- C. Drop the original protections to speed up the model training pipeline, since the new platform applies its own defaults
- D. Delete the audit trail once the data has been transformed
Answer: A
Data protection principles applied at the original location must persist when data moves to the AI solution. Dropping protections or sharing broadly increases risk.
Domain 2: AI Risk Management
Q342 Who is ultimately and PRIMARILY responsible for establishing the organization's risk appetite and tolerance criteria?
- A. Individual developers
- B. The help desk
- C. Owners and/or the governing body
- D. External vendors
Answer: C
The organization's owners and governing body are ultimately responsible for establishing risk appetite and tolerance criteria. Management then determines acceptable levels in line with them. Help desks, vendors and developers do not set appetite.
Domain 2: AI Risk Management
Q343 Acceptable levels of risk are usually defined by management PRIMARILY in terms of what?
- A. The vendor's recommended default settings for the AI platform, as published in the standard installation guide
- B. The number of employees who use the AI system daily
- C. The number of parameters in the deployed model
- D. The losses the organization is willing to accept in pursuit of its goals
Answer: D
Management normally defines acceptable risk by the losses the organization is willing to accept to achieve its goals, informed by appetite and tolerance. Vendor settings or model size do not define acceptability.
Domain 2: AI Risk Management
Q344 Which factor MOST directly constrains acceptable AI risk limits for a healthcare organization?
- A. Compliance requirements such as HIPAA and GDPR
- B. Seating plans for the staff who operate clinical systems
- C. Brand color guidelines applied to patient-facing applications
- D. Office layout decisions affecting where AI teams are seated
Answer: A
Compliance requirements, such as the EU AI Act, HIPAA and GDPR, compel conformance with obligations and shape acceptable limits. Seating, layout and colors do not.
Domain 2: AI Risk Management
Q345 Which description BEST matches appropriate risk as defined in the manual?
- A. Risk an enterprise takes to deliver services, achieve goals and comply with regulation
- B. Risk that is deliberately hidden from the board and regulators
- C. Risk that has no clear benefit, does not align with the organization's goals and is therefore always unacceptable to the board
- D. Risk that exceeds the defined tolerance for the activity
Answer: A
Appropriate risk is the risk necessary to deliver services, achieve goals and comply with regulatory requirements. Inappropriate risk does not align with goals, yields no clear benefit or exceeds the defined limits.
Domain 2: AI Risk Management
Q346 A residual AI risk is within appetite and the cost of further treatment exceeds the benefit. Which response is MOST appropriate?
- A. Transfer
- B. Avoid
- C. Accept
- D. Mitigate
Answer: C
Accepting is appropriate when risk is within acceptable limits and further treatment is not cost-justified. Avoiding would stop the activity, transferring shifts impact, and mitigating adds controls that are not cost-justified here.
Domain 2: AI Risk Management
Q347 An organization decides not to launch an AI product because the risk exceeds tolerance. Which response does this BEST represent?
- A. Avoid
- B. Accept
- C. Transfer
- D. Share
Answer: A
Avoidance means not proceeding with, or discontinuing, the activity that creates the risk. Accepting retains it, and transferring or sharing moves part of the impact to another party.
Domain 2: AI Risk Management
Q348 A team adds human review and input validation to reduce the likelihood of a harmful AI output. Which response does this BEST represent?
- A. Accept
- B. Mitigate
- C. Avoid
- D. Transfer
Answer: B
Mitigation reduces likelihood or impact through controls such as human oversight and validation. Avoid would end the activity and accept retains risk without treatment. Transfer moves the impact.
Domain 2: AI Risk Management
Q349 An organization buys cyber insurance to cover losses from an AI incident. Which response does this BEST represent?
- A. Accept
- B. Transfer
- C. Mitigate
- D. Avoid
Answer: B
Insurance transfers or shares the financial impact of risk with a third party. It does not reduce likelihood, so it is not mitigation. Avoid and accept do not describe buying cover.
Domain 2: AI Risk Management
Q350 A contract assigns part of the liability for AI errors to the vendor. What does this BEST represent?
- A. Avoiding the risk
- B. Transfer or share risk
- C. Eliminating risk
- D. Accepting all risk
Answer: B
Allocating liability by contract shares or transfers risk, but the organization usually retains accountability for outcomes. It does not avoid or eliminate the risk. Accepting all risk is the opposite.