Online Access Free AAIR Exam Questions

A risk practitioner is performing a business impact analysis (BIA) of potential service disruptions to an organization's AI-powered operational processes. Which of the following is MOST important to address in the BIA?

• A.General risk classification criteria and effective replacement cost of AI assets
• B.Effects of AI system outages on core operations and associated recovery time targets
• C.AI development life cycle records and evidence of change control in place
• D.Incident response procedures that account for the tendency of AI incidents to scale quickly

An organization seeks to improve AI oversight by incorporating external perspectives regarding potential public harms from AI use. Which of the following is MOST important to consider?

• A.Quantitative metrics to measure and report fairness and performance
• B.Public policy research checklists that assess potential public impacts
• C.Post-deployment analysis and escalation of observed harms from operational logs and user complaints
• D.Formal procedures to solicit and prioritize feedback from affected communities and independent experts

Which of the following should be the MOST important area of focus during the development of data security risk scenarios specific to AI?

• A.Quantum encryption methods for the protection of proprietary organizational data assets
• B.Development and communication of need-based access policies for the use of AI applications
• C.Documentation of business unit readiness for secure adoption of AI for general operations
• D.Attack vectors enabled by techniques for malicious alteration of AI system outputs

Which of the following is the GREATEST societal risk posed by the dissemination of deepfakes created with AI image generation technology?

• A.Job loss and economic destabilization
• B.Disclosure of personal data
• C.Intellectual property infringement
• D.Widespread misinformation

A risk practitioner is evaluating training datasets for a new AI model. Which of the following approaches BEST reduces fairness risk during model development?

• A.Selection of a single provider for large volumes of training data to reduce variance
• B.Representative sampling strategies combined with bias mitigation controls
• C.Identification and resolution of data labeling inconsistencies
• D.Use of synthetic data to address training dataset imbalances