AI agents are quickly becoming a new interface for commerce. As purchasing shifts from traditional web flows to LLM-driven interactions, developers need practical ways to enable secure, reliable agent-initiated transactions.

This talk explores how to implement agent-driven commerce using emerging protocols and permission-based payment token models.

We’ll focus on real implementation patterns, security boundaries, and UX considerations required to make agentic payments predictable and trustworthy.

Attendees will leave with a clear understanding of how agent-driven transactions work under the hood and how to design systems that support AI-native purchasing experiences.