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 1: AI Governance and Program Management
Q251 Customers raise concern that an AI-driven recommendation process may disadvantage certain groups. Which external stakeholder consideration does this BEST represent?
- A. Preventing discrimination or bias in AI outcomes
- B. Engaging in dialogue to shape fair and realistic AI policies
- C. Reducing the energy consumption of AI systems for sustainability
- D. Ensuring AI systems integrate with partner and vendor platforms
Answer: A
For customers and end users, the considerations include protecting data, explaining AI impact and preventing discrimination or bias. Integration and regulator dialogue concern other stakeholders. Energy use concerns society.
Domain 1: AI Governance and Program Management
Q252 An AI charter is being drafted for a new initiative. What is the PRIMARY purpose of the charter?
- A. To replace the AI policy with a shorter set of operating guidelines
- B. To list approved disciplinary actions for employees who misuse AI tools
- C. To record the final opinion of the audit committee on the AI program results for the previous financial year
- D. To define the scope, objectives and work of the AI initiative and align stakeholders
Answer: D
The charter formally defines scope, objectives and work and aligns stakeholders and expectations. It is not an audit report or policy substitute. Disciplinary matters are unrelated.
Domain 1: AI Governance and Program Management
Q253 A company wants a certifiable standard for an AI management system focused on ethical use, transparency and accountability. Which is the BEST choice?
- A. ISO/IEC 42001
- B. ISO/IEC 23053
- C. IEEE 7000-2021
- D. OECD AI Principles
Answer: A
ISO/IEC 42001 is a voluntary standard for an AI management system, with requirements to establish, implement, maintain and improve it. IEEE 7000 embeds human values in design and ISO/IEC 23053 describes a generic ML framework. OECD principles are recommendations, not a management system.
Domain 1: AI Governance and Program Management
Q254 Developers want a model process to address ethical concerns during system design, including privacy, transparency and fairness. Which resource BEST fits?
- A. EU AI Act
- B. ISO/IEC 42001
- C. NIST AI RMF
- D. IEEE 7000-2021
Answer: D
IEEE 7000-2021 focuses on embedding human values into AI during design across the life cycle. The NIST framework addresses risk management, and the EU AI Act is regulation. ISO 42001 is a management system standard.
Domain 1: AI Governance and Program Management
Q255 Which framework, developed by the US National Institute of Standards and Technology, BEST helps identify, assess and mitigate AI system risk?
- A. NIST AI Risk Management Framework
- B. OECD AI Principles
- C. ISO/IEC 23053
- D. Model AI Governance Framework
Answer: A
The NIST AI RMF assists organizations in identifying, assessing and mitigating AI risk. The Model AI Governance Framework is from Singapore's PDPC. The OECD principles and ISO 23053 are separate efforts.
Domain 1: AI Governance and Program Management
Q256 An organization operating in Singapore asks which framework helps deploy AI responsibly. Which is MOST relevant?
- A. The Model AI Governance Framework from the PDPC
- B. The OECD AI Principles adopted across member countries
- C. The Colorado artificial intelligence consumer protection act
- D. The IEEE 7000 model process for ethical system design
Answer: A
Singapore's PDPC developed the Model AI Governance Framework to help organizations deploy AI responsibly. IEEE 7000 is a design standard and the others are US state laws. Jurisdiction guides relevance.
Domain 1: AI Governance and Program Management
Q257 A compliance officer notes that AI laws are largely untested in litigation. What is the GREATEST implication for the enterprise?
- A. AI use is automatically prohibited until courts rule on the statutes
- B. Compliance obligations may change as laws are drafted and interpreted
- C. Voluntary standards automatically become mandatory for all enterprises
- D. Penalties no longer apply because the rules are still unproven in court
Answer: B
Laws are new and many remain in draft, so obligations and interpretation can shift. Penalties still apply under enacted laws, and voluntary standards do not become mandatory by default. Monitoring change is essential.
Domain 1: AI Governance and Program Management
Q258 Which penalty level is the HIGHEST that applies to the most serious EU AI Act violations?
- A. Up to USD 100 thousand per violation with daily fines continuing
- B. No fines, since certification under the act is voluntary only
- C. Up to EUR 35 million or 7% of global annual revenue
- D. Up to three years of imprisonment for the responsible officer
Answer: C
The EU AI Act sets fines up to EUR 35 million or 7% of global annual revenue, whichever is higher. ISO 42001 has no fines, and state laws have smaller penalties. Imprisonment is cited for South Korea's act in different circumstances.
Domain 1: AI Governance and Program Management
Q259 An analyst finds a language model gives an outdated answer about a recent event. Which limitation does this BEST represent?
- A. Response constraints blocking restricted content categories
- B. Human bias introduced through the way the prompt was written
- C. Token limits truncating the earlier part of the conversation
- D. Lack of context beyond the training data
Answer: D
AI lacks real-world knowledge beyond its training data, so questions about recent events may yield outdated answers. Token limits affect long conversations, and constraints limit allowed content. The scenario is about missing knowledge.
Domain 1: AI Governance and Program Management
Q260 In a lengthy chat, an assistant forgets details stated earlier. Which limitation is MOST likely responsible?
- A. Model drift
- B. Excessive explainability
- C. Data poisoning
- D. Token limit
Answer: D
Reaching the token limit causes the model to forget previously discussed information and lose context. Drift occurs over time with changing data, and poisoning is an attack. Explainability is unrelated.