Trusting the Agent? Security by Design for Agentic AI
Agentic AI extends the capabilities of AI systems beyond generating content to taking autonomous actions and making decisions. This creates entirely new attack vectors and security challenges. In this session, you will learn about the key risks associated with Agentic AI and discover the Security-by-Design principles organizations should adopt to effectively secure and control their AI agents.
Topic
When & Where
Details
Format:
Technology lecture
Language:
German
Session description
Speaker: Pablo Schmücker, Justin Szczepaniak
Agentic AI represents the next evolution of artificial intelligence. Unlike traditional AI systems that primarily generate content or respond to queries, AI agents can make autonomous decisions, execute actions, interact with external tools and systems, and complete complex multi-step tasks with minimal human involvement. While these capabilities unlock significant opportunities for productivity and automation, they also introduce entirely new attack surfaces and security challenges.
From prompt injection and tool abuse to excessive permissions, data exposure, and unintended agent behavior, organizations must address a new class of risks that arise when AI systems move beyond recommendations and begin taking actions on behal ...
