The potential for an AI system to cause harm due to unethical behavior, including bias, discrimination, or violation of privacy.
A subset of AI risk focused on moral and societal harms - harm to dignity, autonomy, fairness, and rights. Ethical risk management involves identifying scenarios (e.g., discriminatory lending), scoring their severity, incorporating them into enterprise risk registers, and implementing controls (e.g., bias checks, consent flows) to keep ethical exposures within acceptable bounds.
A credit-card promotions AI is found to offer higher-interest offers to lower-income neighborhoods. The bank’s ethical-risk register flags this, triggers an investigation, and leads to recalibrating the promotion algorithm to ensure equitable rates regardless of income bracket.
