Welcome to "AI or Not," the podcast where we explore the intersection of digital transformation and real-world wisdom, hosted by the accomplished Pamela Isom. With over 25 years of experience guiding leaders in corporate, public, and private sectors, Pamela, the CEO and Founder of IsAdvice & Consulting LLC, is a veteran in successfully navigating the complex realms of artificial intelligence, innovation, cyber issues, governance, data management, and ethical decision-making.

An AI agent can draft a memo, pull files, trigger actions, and coordinate other agents, so the moment you move beyond a chatbot, you’re also stepping into a new governance and security reality. We’re joined by Nelli Babayan, US Federal AI Director at Microsoft, to break down how federal agencies are adopting agentic AI and what it takes to do it in a way that’s mission-first, compliant, and built to scale.

We walk through a clear three-phase transformation: starting with a human plus an AI assistant, moving to a human working alongside multiple agents, and reaching multi-agent systems that run sequences of tasks under human direction. Along the way, we talk practical federal AI use cases, from case management and information retrieval to science and energy work inside the Department of Energy ecosystem, including national labs exploring everything from lithium battery efficiency to faster nuclear permitting and licensing.

The heart of the conversation is AI governance and AI security. Nelli explains why “an agent without a governed identity is an unmanaged insider,” and what “governed identity” really means in day-to-day operations: observability into which agents exist, who built them, what data they can access, and how users interact with them. We also connect the dots to data governance, shadow AI, and why security and compliance by design can’t be bolted on after a pilot.

If you’re building an AI adoption strategy for government or regulated industries, this is your roadmap to move fast without sacrificing accountability. Subscribe, share this with a colleague working on federal AI, and leave a review if it helps. Then, tell us: what’s the first workflow you want to safely hand to an AI agent?