FlagThis

Abstract

This talk examines the transition of artificial intelligence from experimental hype to operational reality, focusing on the security implications of autonomous AI agents. As organizations deploy agentic AI, they face significant risks regarding governance, non-human identity management, and the lack of "self-awareness" in autonomous systems. The discussion highlights the critical need for security controls to limit the "blast radius" of AI actions and provides insights into recent threat trends, including IoT botnet operations and nation-state psychological hacking.