Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a thought-provoking conference talk from DEF CON 32 that examines the critical challenges surrounding vulnerability disclosure in AI systems. Delve into the complex interplay between transparency and security while understanding how data bias exploitation and model manipulation can magnify vulnerability impacts. Through an analysis of real-world examples and recent disclosures, gain valuable insights into responsible vulnerability management practices for AI systems. Learn how the industry can work together to refine disclosure processes, ultimately strengthening AI system integrity and building public confidence in artificial intelligence technologies. In this 31-minute presentation, discover practical approaches for addressing biases and misinformation that often obscure the true nature of AI vulnerabilities.
Syllabus
DEF CON 32 - Got 99 problems but prompt injection ain't watermelon - Chloé Messdaghi Kasimir Shulz
Taught by
DEFCONConference