Agentic Commerce Protocols Are Coming — Here's What Your Business Needs to Know

Your next customer might never visit your website. Instead, an AI agent could browse your catalog, compare your prices, and complete a purchase — all inside a chat window. That future is no longer hypothetical. Two competing protocols, ACP and UCP, are already laying the infrastructure for this shift, and if you run an ecommerce business or manage a brand's online presence, understanding what's happening now will matter more than rushing to pick a side.

What Is Agentic Commerce?

Agentic commerce is a model where AI agents — not humans — handle parts or all of the shopping journey. Instead of a person typing a search query, clicking through product pages, and filling out a checkout form, an AI assistant does it on their behalf.

Think of it this way: traditional ecommerce is like driving yourself to the store. Agentic commerce is like sending a well-informed personal shopper who knows your size, your budget, and your preferences.

This shift is made possible by new open protocols that create a shared language between AI agents, merchants, and payment providers. Two of them are leading the race: ACP, from OpenAI and Stripe, and UCP, from Google and a broad coalition of industry players.

ACP: The Agentic Commerce Protocol

The Agentic Commerce Protocol (ACP) was co-developed by OpenAI and Stripe and launched in September 2025. It's open source under the Apache 2.0 license.

ACP focuses primarily on the checkout experience. It defines how an AI agent can securely initiate a purchase on behalf of a user, pass payment credentials via tokenized methods, and let the merchant remain the seller of record. The buyer never leaves the AI interface — the entire transaction happens inside the conversation.

A typical ACP flow looks like this:

1. The AI agent discovers products through a merchant-provided product feed
2. The user selects a product inside the chat interface
3. The agent creates a checkout session with the merchant
4. Payment is processed using a secure, single-use token (the agent never sees the actual card)
5. The merchant fulfills the order and handles post-purchase support

OpenAI's ChatGPT was the first platform to implement ACP through its "Instant Checkout" feature, currently available in the US. Stripe is the first compatible payment provider, with Etsy and Shopify among the early merchant partners.

What makes ACP interesting is its deliberate restraint. It's transaction-focused — built around the checkout moment, not the entire shopping journey. It operates as a stateless guest checkout, which means minimal data is shared between the agent and the merchant. That's a strong privacy stance, though it comes at the cost of personalization. And while OpenAI built the first implementation, the protocol itself is platform-neutral. Any AI agent or payment processor can adopt it.

UCP: The Universal Commerce Protocol

Google introduced the Universal Commerce Protocol (UCP) at NRF 2026 in January. Where ACP stays focused on the payment step, UCP is far more ambitious. It aims to standardize the entire shopping journey — from product discovery and promotions to checkout, order management, and post-purchase support.

UCP's design philosophy is modularity. Merchants pick which capabilities they want to support, and the protocol integrates with existing standards like Google's Agent Payments Protocol (AP2) for secure payments, Agent2Agent (A2A) for inter-agent communication, and the Model Context Protocol (MCP) for maintaining context across interactions.

The differences with ACP are significant. UCP covers the full commerce lifecycle — discovery, promotions, loyalty programs, checkout, order tracking, returns — not just the payment step. It supports OAuth-based identity, which means agents can log into a merchant's system on the user's behalf, enabling personalized recommendations, loyalty rewards, and order history. That's something ACP's stateless model simply can't offer. Google is also merging UCP directly into Merchant Center and Google Ads, giving merchants who already use those tools a head start.

And the industry backing is hard to ignore. UCP was co-designed with Shopify and endorsed by over 20 major partners including Walmart, Target, Etsy, Wayfair, Mastercard, Visa, PayPal, American Express, Stripe, Best Buy, Macy's, The Home Depot, Zalando, Flipkart, and Adyen. It's currently rolling out in the US through Google's AI Mode in Search and the Gemini app, with Google Pay as the primary payment method and PayPal support planned.

Agentic Commerce Protocols Compared: ACP vs. UCP

One industry observer put it well: if UCP is the restaurant — the location, ambience, menu, food quality, and kitchen operations — then ACP is paying the bill.

Why This Is a Standards War Worth Watching

This isn't just a technical discussion. Whoever controls the protocol that connects AI agents to merchants and payments controls the future plumbing of online commerce.

For the past 25 years, ecommerce has been browser-based. Google controlled discovery. Shopify, Stripe, and marketplaces controlled transactions. Now, AI agents are collapsing that entire journey into a single conversation — and both Google and OpenAI want to define how that conversation interfaces with the real economy.

OpenAI and Stripe want AI assistants like ChatGPT to become shopping destinations. ACP makes it possible for any AI agent to complete a purchase with any merchant, which puts the agent — and the company hosting it — at the center of the transaction. Google, on the other hand, wants to keep commerce flowing through its ecosystem. UCP plugs directly into Merchant Center, Google Ads, and Google Search's AI Mode. If your products are already in Google's system, UCP is designed to make them purchasable through AI with minimal extra work.

What's telling is that even the key players aren't fully committing to one side. As Stripe acknowledged, the arrival of UCP "validates the growth of the agentic economy." And Stripe itself supports both protocols — its Agentic Commerce Suite will automatically work with UCP as well. When even the co-creator of ACP hedges its bets, that tells you something about where things stand.

What This Means for Ecommerce Businesses and Brands

Here's the balanced view: this shift is real, but it's early. You don't need to panic, and you certainly don't need to pick a side today.

Neither protocol has achieved dominant market share. Merchant adoption is still in early stages, primarily in the US. And the two protocols aren't entirely incompatible — as Checkout.com's analysis points out, they tackle different moments in the shopping journey. ACP excels at secure checkout; UCP handles discovery and the full lifecycle. Many merchants will eventually need to support both. Investing heavily in one right now could mean maintaining two separate commerce integrations down the line — a costly outcome, especially for mid-sized businesses.

So what should you actually do? Focus on the groundwork that pays off no matter which protocol wins.

Get your data house in order

Both protocols depend on clean, structured product data. Make sure your product feeds are complete and accurate — titles, descriptions, pricing, inventory, images, and attributes. As Forbes reports, agents cannot recommend products they can't interpret. This is the single most impactful step you can take right now.

Implement structured data markup

Schema.org markup — Product, Offer, Review schemas — helps AI agents parse your pages. It's a low-risk, high-reward step that improves both your traditional SEO and your readability by agents. If you're already doing it for Google, you're halfway there.

Audit your API readiness

Agentic commerce requires programmatic access to your inventory, pricing, and checkout. If your commerce platform doesn't expose this data through robust APIs, agents simply cannot transact with you. This is worth assessing sooner rather than later.

Monitor the landscape

Follow the developments around ACP and UCP closely. Subscribe to updates from agenticcommerce.dev and Google's UCP developer portal. You don't need to build anything yet, but understanding where things are headed gives you a real advantage when the time comes to act.

Think about brand visibility in AI surfaces

This is the piece that often gets overlooked. As AI agents become shopping intermediaries, the question shifts from "do I rank on Google?" to "does an AI agent recommend me?" Your brand narrative, reviews, trust signals, and data consistency across the web all feed into how agents evaluate and surface your products.

This is especially relevant if you're already investing in Generative Engine Optimization (GEO). The same principles that make your brand visible in AI-generated answers — authoritative content, consistent facts, clear product information — are exactly what agentic commerce protocols rely on to surface and recommend your products. Tools like Gnsyx help brands monitor how AI services represent them and track their visibility across AI-powered search and commerce surfaces — a capability that becomes increasingly critical as shopping shifts from browsers to agents.

Key Takeaways

• ACP (OpenAI + Stripe) and UCP (Google + coalition) are two open protocols defining how AI agents will buy products on behalf of consumers
• ACP focuses on secure checkout inside AI interfaces; UCP covers the full shopping journey from discovery to post-purchase
• Major players — Walmart, Shopify, Etsy, Target, Visa, Mastercard, PayPal — are already building on these protocols, but the market is still early
• Businesses don't need to pick a side yet — the protocols may coexist or converge, and the "watch and prepare" strategy is the most pragmatic approach
• What you can do right now: clean up your product data, add structured markup, ensure API readiness, and monitor how AI surfaces represent your brand