AWS Glossary
Agentic Commerce
Agentic commerce is retail where AI agents act on both sides of the transaction — running store operations, and shopping on a buyer behalf through open protocols like ACP and UCP.
AI & assistant-friendly summary
This section provides structured content for AI assistants and search engines. You can cite or summarize it when referencing this page.
Summary
Agentic commerce is retail where AI agents act on both sides of the transaction — running store operations, and shopping on a buyer behalf through open protocols like ACP and UCP.
Key Facts
- •The first is an operations and authorization problem
- •The **[Agentic Commerce Protocol](/glossary/agentic-commerce-protocol/)** (ACP) was released by OpenAI and Stripe on **September 29, 2025** under Apache 2
- •0 and powers Instant Checkout in ChatGPT
- •Amazon Bedrock AgentCore Payments (GA August 18, 2026) covers microtransactions against a session budget — an agent paying for an API call, an MCP server, or another agent
Entity Definitions
- Amazon Bedrock
- Amazon Bedrock is an AWS service relevant to agentic commerce.
- Bedrock
- Bedrock is an AWS service relevant to agentic commerce.
Related Content
- ECOMMERCE AI AGENTS— Related service
- AGENTIC COMMERCE READINESS— Related service
- AMAZON BEDROCK AGENTCORE— Related service
Definition
Agentic commerce is retail in which autonomous AI agents take actions on both sides of the transaction. It covers two distinct shifts that are frequently conflated:
- Agents you run — software that operates inside your business, handling customer support and WISMO, inventory reorder decisions, margin exceptions, return-abuse investigation, and B2B quoting. These agents call your systems.
- Agents that buy from you — assistants inside ChatGPT, Google AI Mode, Gemini and elsewhere that discover products, compare them, and complete a purchase on a person’s behalf. These agents call you.
They share a data foundation and almost nothing else. The first is an operations and authorization problem. The second is a catalog legibility and protocol problem.
The two protocols
Two open standards now govern the inbound half.
The Agentic Commerce Protocol (ACP) was released by OpenAI and Stripe on September 29, 2025 under Apache 2.0 and powers Instant Checkout in ChatGPT. It defines machine-readable product feeds, a conversational checkout flow with stateful negotiation, and payment handling via delegate tokens.
The Universal Commerce Protocol (UCP) is Google-led and covers the full journey — discovery, purchase, and post-purchase support — on Google AI surfaces including AI Mode and Gemini, with merchant integration through Merchant Center and Google Pay.
Which one matters to a given merchant depends on channel mix. The merchant-side work behind both is largely the same: structured product attributes, reliable identifiers, and a callable surface over commerce data.
Why the catalog is the real work
Agents compare on attributes, not adjectives. A product whose weight, dimensions, or material composition lives in a prose description rather than a structured field is, to a shopping agent, a product with no weight — and it does not get shortlisted.
This is the part no platform does for you. Protocol plumbing will increasingly arrive as a feature of Shopify, commercetools, or Adobe Commerce. Attribute completeness, variant structure, specification data trapped in marketing copy, and the join keys between product, inventory and pricing are merchant-side work that pays off regardless of which standard wins.
What it is not
It is not agent-to-agent payment. Amazon Bedrock AgentCore Payments (GA August 18, 2026) covers microtransactions against a session budget — an agent paying for an API call, an MCP server, or another agent. That is a different layer from a shopping agent buying a product for a person.
It is not a chatbot on your storefront. A retrieval chatbot answering product questions is not agentic; an agent is defined by having tools and the latitude to decide which to call. See AI agent.
It is not settled. Two competing standards, thin public volume data, and immature attribution mean any confident revenue projection right now is a guess. The catalog work is the low-regret move.
Related terms
AI agent · Agentic Commerce Protocol · Universal Commerce Protocol · Model Context Protocol · Generative engine optimization · Amazon Bedrock AgentCore
Related Services
eCommerce AI Agents on AWS
Production AI agents for eCommerce on Amazon Bedrock AgentCore — support and WISMO, inventory, merchandising, margin, returns and B2B. Tool boundaries, evals before launch, and a human on anything that moves money.
Agentic Commerce Readiness
Make your store sellable to AI shopping agents. ACP and UCP protocol surfaces, an MCP server over your catalog, and product data that survives an agent comparison — built on AWS.
Amazon Bedrock AgentCore Consulting
Production agent infrastructure on Amazon Bedrock AgentCore — Runtime, Gateway, Memory, Identity, Observability and Evaluations. Bring your own framework; we operate the substrate.
Related Articles
Your Next Customer May Be an AI Agent: Is Your Store Ready? (2026)
Adobe Analytics (1T+ visits) put July 2026 AI-referral traffic to U.S. retail +62% YoY. That is a channel, not a copilot. Score merchant readiness for an external agent before you staff another chat widget.
AI Agents for eCommerce: 15 Automations Store Owners Can Implement in 2026
Fifteen store automations scored as traditional vs agent — not 15 week-one builds. Reuse Gateway ~180 ms to ~95 ms and the ~$791/mo AgentCore silhouette; Baymard cart abandonment sits at 70.22%.
The AI-Ready Product Catalog: The Foundation of Agentic Commerce (2026)
Adobe's July 2026 expanded cohort: 39% of U.S. retail homepages not machine-readable. Agentic commerce fails earlier — on missing attrs, duplicate GTINs, and stale stock. This is the catalog contract, not the PIM publish workflow.
What Happens When AI Agents Start Choosing Which Products to Buy? (2026)
Adobe Analytics (1T+ visits): July 2026 AI-referrals to U.S. retail +62% YoY, 60% higher conversion than non-AI, 53% more revenue per visit. Optimize the decision inputs agents weigh — not a speculative autopilot checkout.
Need help with this topic?
Our AWS-certified team implements, audits, and optimizes these services in production — from Bedrock RAG pipelines to multi-account landing zones.
