EP 536: Agentic AI - The risks and how to tackle them responsibly

In this episode of the Everyday AI Show, host Jordan Wilson interviews Sarah Bird, Chief Product Officer of Responsible AI at Microsoft, about the evolving landscape of responsible AI in the age of agentic and multi-agentic systems. They discuss how the increasing capabilities of AI agents, especially in task completion and coordination, introduce new risks and challenges related to ethics, governance, and security. Sarah emphasizes the importance of adapting security systems to manage AI agents, testing agent systems thoroughly, and monitoring their performance to ensure they stay on task and don't leak sensitive data. She also highlights the shift from human-in-the-inner-loop to human-in-the-outer-loop, where humans focus on pre-deployment testing and post-deployment monitoring. The conversation also covers the need for workforce upskilling to collaborate effectively with AI agents and the critical role of intentional planning and investment in responsible AI practices.