#08 - 05 - AI Explainability Score Card

Notas do episódio

In this episode, we sit down with Michael Novack to talk about the AI Explainability Scorecard, a practical framework for evaluating whether AI explanations are actually useful, trustworthy, and actionable. Michael walks us through the SCORE framework, covering five key dimensions: Sound, Consistent, Optimize, Readable, and Easy. We discuss why “the AI gave an explanation” is not enough, how convincing explanations can still be misleading, and why teams need a structured way to measure explainability before trusting AI-driven decisions. From compliance and auditability to security, governance, and real-world accountability, this conversation explores what it takes to move AI explainability from vague promise to practical discipline. If you are building, securing, reviewing, or approving AI systems, this episode is for you. Become a supporter of this podcast: https://www.spreaker.com/podcast/devsecops-podcast--4179006/support. Apoio: Nova8, Snyk, Conviso, Gold Security, Digitalwolk e PurpleBird Security.