Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Explore the emerging security paradigm needed to protect next-generation AI applications in this 25-minute conference talk from DevSecCon. Discover how agentic AI and AI-native applications are expanding attack surfaces and introducing novel risks including prompt injection, toxic flows, and unpredictable non-deterministic behavior. Learn why traditional rules-based security tools designed for deterministic systems are inadequate for securing modern AI applications. Understand the need for a new breed of security professionals - AI Security Engineers - who must leverage their own agentic security orchestration systems to enable secure AI innovation. Examine the core tasks these orchestration systems must cover and the opportunities they present for identifying, understanding, and addressing complex threats against AI-native applications.
