AI can answer a question. But what happens when AI starts making decisions and taking action on your behalf?

On this episode of The Peel, Shelley McGuire sits down with Jason Balser, Senior Director of AI and Data Strategy at RELI Group, to break down one of the hottest topics in technology today: agentic AI. Jason explains the concept through a simple travel analogy—traditional AI can help with individual pieces of planning a trip, while agentic AI is more like handing the entire itinerary to a travel agent and giving them the autonomy to make decisions to accomplish your goal.

But with greater autonomy comes greater responsibility. Jason explores what happens when AI agents make unexpected decisions, why hallucinations and weak controls can create significant risks, and why organizations need the right governance and guardrails in place before turning over too much control.

The conversation also tackles a critical question every leader should ask before deploying an AI agent: What’s the worst thing that could go wrong? From managing an email inbox to decisions affecting healthcare or national security, not every use case carries the same consequences—and not every decision should be automated.

Jason shares three considerations for responsible agentic AI: define the outcome, understand the consequences, and make sure a real person remains accountable. As agentic AI evolves, the challenge won’t simply be adopting it—it will be determining where autonomy makes sense and building governance that matches the risk.

To connect with Jason, follow him on LinkedIn here.
Learn more about RELI Group at their website here.