S.A.F.E. Artificial Intelligence in Finance
Conference
65th ISI World Statistics Congress 2025
Format: IPS Abstract - WSC 2025
Keywords: accuracy, algorithmic-fairness, explainable ai, machine learning, robustness
Session: IPS 1028 - Financial Data Science: Opportunity and Risks
Monday 6 October 2 p.m. - 3:40 p.m. (Europe/Amsterdam)
Abstract
The growth of Artificial Intelligence (AI) applications in finance requires to develop risk management models. In this paper, we contribute to the debate on regulations and industry standards for AI risk management models proposing a set of integrated safe AI statistical metrics. Our proposed metrics are consistent with each other, as they are all derived from a common underlying statistical methodology: the Lorenz curve. They are easy to interpret, as are all expressed in percentages of an ideal situation of full compliance. They are agnostic, as they can be applied to any machine learning method. They are fully reproducible, using the proposed Python code. Our methodological framework, named “Rank Graduation Box”, allows to derive any necessary compliance metric for AI using a pairwise comparison of Lorenz curves, based on different rank graduations. We specifically consider metrics for the assessment of Sustainability, Accuracy, Fairness, Explainability, and Privacy. Our proposal allows the application of all the proposed metrics for the assessment of the regulatory compliance of any AI applications in finance.