AI Risk
The potential for AI systems to cause harm or unintended consequences, including ethical, legal, and operational risks.
The spectrum of potential negative outcomes - technical failures, unfair treatment, privacy breaches, legal violations, reputational damage - quantified qualitatively or quantitatively to guide governance decisions.
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.
