The AAIA™ Study Guide emphasizes that regular algorithmic audits are critical for identifying unintended consequences such as the promotion of harmful or biased content. This proactive approach helps maintain trust and ensure that algorithmic decisions align with organizational values and ethical standards.
"Auditing and monitoring AI models regularly helps detect and correct bias, drift, or other unintended behavior. It is essential for high-impact AI systems like content recommendation engines." While content customization (A) and user consent (C) are helpful, they don't prevent bias propagation.
Suspension (B) may halt engagement and isn't sustainable. Therefore, D is the most balanced and strategic solution.
Reference: ISACA Advanced in AI Audit™ (AAIA™) Study Guide, Section: "AI Governance and Risk Management," Subsection: "Model Monitoring and Risk Mitigation Strategies"