Google has unveiled an Apache 2–licensed open standard to let AI agents autonomously handle shopping, checkout, and payments, aiming to eliminate custom integrations and reduce cart abandonment.
Google has introduced the Universal Commerce Protocol (UCP), a new open standard designed to enable AI agents to autonomously perform end-to-end commerce activities, spanning product discovery, purchasing, checkout, payments, and post-purchase experiences.
Released under the Apache 2 open-source licence, UCP is positioned as an openly governed, interoperable framework rather than a proprietary commerce system. Google says the protocol defines the foundational “building blocks for agentic commerce”, allowing AI-driven shopping experiences to operate across platforms without bespoke integrations.
Describing UCP as the “core plumbing for agentic commerce”, Google aims to standardise how AI agents, retailers, platforms, and payment providers transact through a single shared protocol. The company says this approach removes the need for fragmented, custom-built connections that currently slow down digital commerce.
“UCP establishes a common language for agents and systems to operate together across consumer surfaces, businesses and payment providers,” wrote Vidhya Srinivasan, VP & GM for Ads & Commerce at Google, adding that the protocol is compatible with Agent2Agent (A2A), Agent Payments Protocol (AP2), and Model Context Protocol (MCP).
UCP was co-developed with major retail and commerce companies including Shopify, Etsy, Wayfair, Target, and Walmart, signalling broad industry alignment around open agentic commerce standards.
Google plans to deploy UCP through a new checkout feature on eligible product listings in AI Mode in Search and the Gemini app, initially for select U.S. retailers, with international expansion expected. Transactions will use Google Pay and Google Wallet, with PayPal support planned, while retailers remain the seller of record.
Google positions UCP as a key step towards reducing cart abandonment and enabling frictionless, AI-driven shopping at scale.
