Publication Peer Review

This white paper focuses on what we define as an enterprise's "core security responsibilities" around Artificial Intelligence (AI)  and ML (Machine Learning): data security, model security, and vulnerability management. As organizations have duties to uphold for secure and safe AI practices, this white paper and two others in this series provide a blueprint for enterprises to fulfill such organizational responsibilities. Specifically, this white paper synthesizes expert-recommended best practices within core security areas - data protection mechanisms, model vulnerability management, MLOps pipeline hardening, and governance policies for training and deploying AI responsibly. This series's other two white papers discuss additional aspects of secure AI development and deployment for enterprises. By outlining recommendations across these key areas of security and compliance in three targeted white papers, this series aims to guide enterprises in fulfilling their obligations for responsible and secure AI development and deployment.