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
Q321 If AI risk is not effectively managed, what is the MOST likely consequence for the organization?
- A. Harmful bias is introduced and confidence and trust erode
- B. Model performance improves without any monitoring
- C. Training data automatically becomes more accurate
- D. Regulatory obligations are waived
Answer: A
Unmanaged AI systems can quickly introduce harmful biases and erode confidence and trust, which is why risk frameworks emerged. Data does not improve on its own and obligations are not waived. Monitoring is still required.
Domain 2: AI Risk Management
Q322 A risk analyst says AI risk depends on how, where and by whom a system is used. Which principle does this BEST reflect?
- A. Risk is constant across all organizations
- B. Risk applies only to developers
- C. Risk is purely technical
- D. Risk is contextual rather than isolated
Answer: D
The manual states that risk does not exist in isolation and is contextual, so it must be assessed against the organization's use of AI. Treating it as uniform or purely technical ignores usage context and stakeholders.
Domain 2: AI Risk Management
Q323 When a company starts an AI risk assessment, which action BEST helps ensure that risks are identified completely?
- A. Include AI assets from the inventory and consider technical, ethical and legal risk areas
- B. Limit the scope to the models built in-house by the data science team
- C. Use only vendor-supplied risk ratings, without any internal validation
- D. Defer identification until after incidents have shown where the risk lies, and then update the register accordingly
Answer: A
A complete assessment builds on the AI inventory and examines technical, ethical and legal risk across all AI in use. Limiting scope to one team ignores shadow and vendor AI. Vendor ratings and waiting for incidents are reactive.
Domain 2: AI Risk Management
Q324 An organization adopting the NIST AI RMF wants to establish risk policies, accountability structures and a culture of risk management. Which function does this BEST describe?
- A. Measure
- B. Map
- C. Govern
- D. Manage
Answer: C
Govern establishes policies, accountability structures and governance mechanisms that foster a risk management culture. Map identifies system characteristics and impact zones. Measure develops metrics and Manage implements responses.
Domain 2: AI Risk Management
Q325 A team lists an AI system's intended purpose, stakeholders and potential impact zones. Which NIST AI RMF function is it MOST likely performing?
- A. Manage
- B. Measure
- C. Govern
- D. Map
Answer: D
Map identifies AI system characteristics, risk areas and potential impact zones in context. Govern covers policy and accountability. Measure covers metrics and monitoring, and Manage covers responses.
Domain 2: AI Risk Management
Q326 A risk manager develops metrics to monitor AI risk throughout the life cycle. Which NIST AI RMF function does this BEST represent?
- A. Manage
- B. Govern
- C. Measure
- D. Map
Answer: C
Measure develops risk metrics and ways to monitor and manage identified risk through the AI life cycle. Map defines context and Govern sets policy. Manage prioritizes and acts on risk.
Domain 2: AI Risk Management
Q327 After assessing risk, a team prioritizes responses and runs compliance checks in line with potential impact. Which NIST AI RMF function does this BEST describe?
- A. Govern
- B. Manage
- C. Map
- D. Measure
Answer: B
Manage implements risk response strategies and compliance checks and prioritizes action commensurate with impact. Measure produces the information about risk. Map and Govern come earlier.
Domain 2: AI Risk Management
Q328 How does the NIST AI RMF MOST differ from the EU AI Act in classifying risk?
- A. It requires conformity assessments for every AI system an organization uses, whether or not the system is classified as high risk
- B. It is a mandatory regulation applicable worldwide to all AI developers
- C. It is flexible and industry-agnostic without fixed risk categories, while the Act defines explicit categories
- D. It prohibits specified AI uses by law and sets penalties for providers
Answer: C
NIST provides flexible, voluntary tools without rigid thresholds, whereas the EU AI Act sets explicit risk classes and mandatory measures. Conformity assessments and prohibitions belong to the Act. NIST is voluntary.
Domain 2: AI Risk Management
Q329 Which statement about the EU AI Act is MOST accurate for a non-EU company?
- A. It is exempt because it is based outside the EU
- B. It must comply if its AI systems affect people in the EU
- C. It need only comply for unacceptable-risk systems
- D. It may ignore the Act if it uses a third-party provider
Answer: B
The Act applies extraterritorially, so organizations outside the EU must comply when their AI systems impact EU citizens. Location of incorporation and outsourcing do not remove obligations.
Domain 2: AI Risk Management
Q330 Which set of categories BEST represents the four risk levels used in the EU AI Act?
- A. Govern, map, measure and manage
- B. Systemic, statistical, human and emergent
- C. Unacceptable, high, limited and minimal
- D. Critical, serious, moderate and low
Answer: C
The Act classifies AI systems as unacceptable, high, limited or minimal risk. The other sets are not its categories; the last is the NIST AI RMF functions and the third lists NIST bias types.