Overview
Explore the vulnerabilities of AI systems in this 21-minute conference talk from RSA Conference. Delve into cyberthreats like dolphin attacks and kangaroo-induced confusion in self-driving cars. Gain insights from Itsik Mantin, Lead Scientist at Imperva, on the security challenges posed by AI technologies. Learn effective strategies for securely implementing learning technology in adversarial environments. Understand the cyber-lifecycle, assess and mitigate AI-related risks, and discover best practices for integrating AI technology securely. Topics covered include security in the hype-cycle, adversarial thinking, blind spots, AI deception threats, and harnessing AI for cybersecurity attack detection.
Syllabus
Intro
Security and the Hype-Cycle
The Australian Challenge
Adversarial Thinking
Blind Spots
Getting Hit by an Ostrich Adversarial Examples
Dolphin Attack
Define Al Deception
Al Deception Threat
Are We Better? HI Deception
Risk Mitigation
Harnessing Al for Cyber Security Attack Detection in Imperva Counter Breach
Summary
Taught by
RSA Conference