OWASP Phoenix Chapter May Meeting
Agentic AI Design Patterns and Their Security Implications
Presenter: Kishan Nagendra
Presentation: This talk explores common agentic AI design patterns and examines how these patterns reshape the security attack surface. It focuses on real-world risks such as delegated identity abuse, cross-agent trust violations, state leakage, and unsafe autonomy boundaries along with remediation strategies.
Agentic AI Design Patterns and Their Security Implications
Presenter: Kishan Nagendra
Presentation: This talk explores common agentic AI design patterns and examines how these patterns reshape the security attack surface. It focuses on real-world risks such as delegated identity abuse, cross-agent trust violations, state leakage, and unsafe autonomy boundaries along with remediation strategies.
Presenter Bio: Kishan is a Security Engineer with six years of software development and application security experience, specializing in AI/ML security. He currently works at Certus Cybersecurity, where he leads AI security research and has helped secure multiple S&P 500 companies. He has also presented at the Phoenix Security & Audit Conference on securing the AI/ML pipeline.
Good to know
Highlights
- 1 hour 30 minutes
- In person
- Free parking
Location
108 W Main St
108 West Main Street
Mesa, AZ 85201
How do you want to get there?
