Very much worth reading this interesting and important interview with Cynthia Rudin, a Duke computer scientist who is working to prove that black box machine learning models "can be transformed into interpretable glass boxes that show their work."

“The whole framework of machine learning just needs be changed when you’re working with something higher-stakes," Rudin told Allison Parshall in @QuantaMagazine: https://www.quantamagazine.org/cynthia-rudin-builds-ai-that-humans-can-understand-20230427/