Build a Learning Habit
Download Class Central's free printable study calendar
Download for Free
Learn how to secure AI systems through threat modeling and design-first security strategies in this conference talk from RSA Conference. Explore why traditional "penetrate and patch" security approaches have failed over 50 years of penetration testing and discover how shifting security left through threat modeling is finally gaining traction. Examine the unique challenges of securing Large Language Models (LLMs) where code and data are intermingled, making traditional security approaches inadequate. Understand the limitations of reactive security measures and why building secure systems remains difficult despite decades of security practices. Discover practical, achievable strategies for delivering AI that is secure by design rather than secured as an afterthought. Gain insights into modern security approaches that move beyond the failed paradigms of the past and learn how to apply threat modeling specifically to AI systems. Master techniques for identifying and addressing security vulnerabilities in AI applications before they reach production, focusing on proactive rather than reactive security measures.
