Agentic AI is now a buzzword in software architecture discussions, but what does it mean for senior practitioners building production systems? In this 60-minute live roundtable, four expert panelists will discuss what makes AI "agentic" and how it differs from traditional AI implementations. When does agentic AI solve problems that conventional approaches cannot? What are the real-world use cases where the added complexity is justified? We'll discuss the security implications of agentic AI, including how to maintain control while enabling automation. Balancing between innovation and practicality, the panel will discuss the "solution in search of a problem" trap and the risks of over-engineering systems with unnecessary autonomous capabilities. Join us for an experience-driven discussion that will help you make informed architectural decisions about when and how to incorporate agentic AI into your deployments.