Run bias detection and fairness analysis on AI models. Store evaluation results, compare metrics over time, and flag disparities.
5 endpoints