4. AI Governance and Responsible AI Leadership (24%)
A hiring team adopts an AI screening tool that ranks thousands of applicants a week. It saves hundreds of recruiter hours and the business case looks excellent. A year later, an analysis shows the tool consistently ranked candidates from certain backgrounds lower, because it learned from a decade of past hiring decisions. Nobody had asked who was accountable for the tool's fairness, no one had tested it across applicant groups, and there was no human review of rejections. The savings were real. So were the legal claims, the regulatory inquiry, and the damage to the employer brand.
Responsible AI and governance are not the brakes on AI value; they are what allow value to survive contact with customers, employees, regulators, and the public. This domain covers three tasks: applying responsible AI principles to business decisions (Task 3.1), establishing governance structures and regulatory compliance (Task 3.2), and identifying enterprise AI risks and directing their mitigation (Task 3.3).
The exam tests these at the level of a business leader. You will not be asked to implement encryption or configure a guardrail. You will be asked who should be at the table, when governance should start, which risks need human oversight, how to scale controls to risk, and how to respond when bias, hallucination, harmful content, or intellectual property concerns appear.
⚠️ Exam Trap: Governance answers on this exam are rarely "stop using AI" or "accept the risk to move fast." The preferred answers are proportionate: classify risk, apply controls that match it, keep humans accountable for high-impact decisions, monitor continuously, and document trade-offs.
