Which aspect is included in CPMAI's risk registers for operations?

• A. Data risk (quality/privacy), model risk (bias, reliability), and operational risk (availability, security, governance).
• B. Market risk, liquidity risk, and credit risk.
• C. Hardware failure risk only.
• D. Reputational risk only.

Answer: (A)

In CPMAI, when you’re looking at risks that affect operations, you focus on three interconnected areas: data risk, model risk, and operational risk. Data risk covers issues with data quality and privacy—trustworthy, accurate data and compliant handling are essential for reliable AI outputs. Model risk addresses biases, unfair or unpredictable behavior, and the model’s reliability over time, including how it might drift or fail under different conditions. Operational risk encompasses how the system runs day to day—availability, security, governance, incident response, access controls, and overall process management to keep the AI system functioning securely and within policy.

These three areas together provide a comprehensive view of operational risks in AI projects, because problems can arise from the data used, the models themselves, or the way the system is operated and governed. The other options focus on risks outside the typical CPMAI operations scope (financial-market risks like market, liquidity, or credit risk; or overly narrow concerns like hardware failure or reputational risk alone), so they don’t capture the full spectrum of operational AI risk as CPMAI defines it.