Sleeping with One AI Open

Explore the world of adversarial attacks against Machine Learning (ML) in this 56-minute conference talk from the 44CON Information Security Conference. Gain a brief introduction to core machine learning concepts before delving into a comprehensive taxonomy of adversarial attacks. Learn about each type of attack in detail, focusing on real-world applications, examples, potential consequences, and mitigation strategies. Presented by Marta Janus, Principal Researcher at HiddenLayer, and Eoin Wickens, Senior Researcher at HiddenLayer, this talk offers valuable insights into the security risks associated with AI/ML systems integrated into personal devices, financial systems, medical equipment, and critical infrastructure. Discover how attackers can abuse these technologies, potentially causing disruption, financial loss, reputational harm, and even risks to human health and life. Benefit from the speakers' extensive experience in cybersecurity, threat intelligence, and malware analysis as they share their expertise on this crucial topic in the evolving landscape of artificial intelligence and machine learning security.

Marta Janus and Eoin Wickens - Sleeping with one AI open