MITRE ATLAS: Exploring AI Vulnerabilities and Attack Vectors

Explore AI system vulnerabilities through a 14-minute conference talk from Security BSides London that examines the MITRE Adversarial Threat Landscape for AI Systems (ATLAS) framework. Gain insights beyond common LLM jailbreaking techniques to understand the broader spectrum of potential attacks targeting AI technologies. Learn through detailed case studies highlighting concerning AI security incidents from recent years, demonstrating the real-world implications of these vulnerabilities as AI adoption continues to accelerate across industries.