10:45 AM - 11:11 AM (PDT)
AI Threats & Vulnerabilities

MITRE’s Jonathan Broadbent and Dr. Christina Liaghati discuss AI Threats & Vulnerabilities, real-world observations, demonstrations from ML red teams and security groups, and the state of the possible from academic research and MITRE ATLAS, the Adversarial Threat Landscape for AI Systems, a collaboratively developed knowledge base of adversary tactics, techniques (MITRE ATLAS™ Introduction - YouTube). They walk through how threats and vulnerabilities have and can continue to impact AI systems, showing real-world examples and talked about MITRE’s growing mechanism for collaboration with ongoing opportunities for industry leaders to shape the future of AI Assurance. If you are interested in driving action as part of the ATLAS Consortium or you would like to learn more about the ATLAS framework, email Jonathan at jbroadbent@mitre.org and Christina at cliaghati@mitre.org and the full MITRE ATLAS team at atlas@mitre.org.

Session Survey