Building the future of agentic commerce: A conversation with Checkout.com’s Rami Josef

Hey there,

Today, I explore one of the most transformative trends reshaping the payments landscape: agentic commerce. Joining me is Rami Josef, Director of Product Management at Checkout.com, to discuss how artificial intelligence is redefining the way consumers shop, and how Checkout.com is helping merchants prepare for this new frontier. Together, we unpack what this means for the future of payments and partnerships across the ecosystem. 

Rami, let’s start with the basics. What exactly do you mean by “agentic commerce”?

For decades, e-commerce has been defined by consumers manually browsing, comparing, and purchasing through websites or apps. But that’s changing fast. Agentic commerce describes a world where AI-powered agents (think intelligent digital assistants) shop on behalf of consumers. 

These agents don’t just react to instructions; they act with intent, making decisions, managing subscriptions, discovering products, and negotiating purchases.

At Checkout.com, we see this as a fundamental shift from human-initiated transactions to AI-delegated ones. But for that shift to work at scale, we need trusted, transparent, and secure infrastructure, and that’s where we’re focusing our efforts.

How is Checkout.com helping to build the foundation for this new type of commerce?

We’re privileged to sit at the intersection of merchants and AI platforms, processing payments for some of the world’s biggest brands. Now we’re extending that expertise into the agentic era.

We’re developing the building blocks that allow merchants and AI agents to transact safely and intelligently. Together with Visa and Mastercard, we’re creating the standards for an agent-powered economy, including:

1. Know Your Agent (KYA) frameworks — giving every AI agent a digital fingerprint so merchants and issuers can identify them.
2. Agentic Credentials Framework — ensuring AI agents have the right, secure credentials for every context (from network tokens to wallet credentials).
3. Biometric-first authentication — offering seamless, phishing-resistant consumer verification.
4. Personalization signals — helping AI agents make smarter decisions based on trusted transaction data (with user consent).
5. Consent and control management — letting consumers define what their AI agents can and can’t do on their behalf.

You mentioned personalization. How does that fit into agentic commerce?

It’s central. Agentic commerce isn’t just about secure transactions; it’s about smarter ones. With consumer consent, Checkout.com can enrich AI platforms with anonymized context from transaction networks.

For example, if an AI agent is booking a hotel, we could provide insights that allow the AI to suggest hotels near their favorite coffee spot. It’s about bridging financial data and decision-making so AI can act more like a personal assistant rather than a basic tool.

One of the challenges with AI in commerce today is reliability. How do you ensure agents operate safely and within boundaries?

That’s a critical question. We believe consumers must stay in control of their agents’ actions. Our consent and control management system will let users define parameters such as:

• “Don’t spend more than $300.”
• “Only buy the Sony headphones if the price drops to $300.”
• “Book flights once per quarter.”

We capture and enforce this consent, creating operational trust for AI platforms and downstream assurance for merchants and issuers. It means transactions are traceable, defensible, and transparent, even if disputes arise.

You mentioned “smarter product discovery.” What does that look like in practice?

Today, AI platforms often rely on virtual cards or web crawling to simulate purchases. Those methods are costly, inefficient, and often unsafe.

Our vision is to replace this with a “network of networks” which we call Inventory Reach. This would let AI agents directly query merchants’ real-time inventory, pricing, and availability. It removes the guesswork and creates a two-way information flow:

• Merchants → Checkout.com → Agents: real-time stock, fulfillment options, and return policies.
• Agents → Checkout.com → Merchants: intent signals, consumer preferences, and loyalty data.

It’s a smarter, more collaborative ecosystem where merchants retain control, consumers gain protection, and AI platforms can scale safely.

What role do merchants play in this vision? Are they passive participants or active collaborators?

They’re absolutely active collaborators. In fact, one of the most exciting frontiers is what we call “guiding agents inside the store”. Merchants will be able to set guardrails and guidance through APIs. For instance:

• Highlight sustainable fashion for a specific customer.
• Request extra ID verification for high-value items.
• Offer premium customers express delivery automatically.

This turns the relationship between merchants and AI agents into a partnership and real-time, API-driven collaboration inside the digital storefront.

And finally, how do you see this space evolving over the next few years?

We’re still in the early stages, but the direction is clear. Today, agents shop with little context, and merchants often have no visibility into their behavior. Tomorrow, we’ll see:

• Contextual, personalized, and dynamic experiences.
• Transparent interactions between merchants and agents.
• A safer and more intelligent commerce ecosystem.

Our mission is to make that evolution possible by securing payments, enabling trust, empowering merchants, and enhancing discovery. In the future, AI won’t just transact, it will collaborate. And Checkout.com is building the rails to make that happen.