Defend & Protect Together: Tools for AI Security & Assurance

MITRE, a not-for-profit (charity organization) serving in the public interest and solving problems for a safer world, in collaboration with 60+ international government, industry, and academic organizations, has built the Adversarial Threat Landscape for Artificial-Intelligence Systems (ATLAS), an open-source framework, tools, and coalition to address current real-world challenges in securing and assuring AI-enabled systems (see atlas.mitre.org). This session will demonstrate how to use ATLAS, a freely available, living knowledge base of adversary tactics, techniques, and case studies for machine learning systems based on real-world observations, demonstrations from machine learning red teams and security groups, and the state of the possible from academic research. With these tools and collaborations, we will lay out actionable recommendations for the global community to understand and mitigate their own security and assurance risks for AI-enabled systems.