AI Risk
The potential for AI systems to cause harm or unintended consequences, including ethical, legal, and operational risks.
Definition
The spectrum of potential negative outcomes—technical failures, unfair treatment, privacy breaches, legal violations, reputational damage—quantified qualitatively or quantitatively to guide governance decisions.
Real-World Example
An insurer quantifies potential losses if its claims-processing AI misclassifies high-value claims as low risk. They allocate additional capital reserves and incorporate manual-review steps for high-value claims, mitigating operational and reputational exposure.
.svg)
