Technology

Amazon’s Secret Plan to Own the World’s Product Data and Fight AI Shopping Agents

admin · · 7 min read

Illustration of Amazon's World Product Graph concept, a vast network of product data connecting everyday items to an AI shopping assistant

Amazon’s Secret Plan to Own the World’s Product Data and Fight AI Shopping Agents

Amazon has a secret plan to make sure shoppers never need to leave its website: own the product information for virtually every item for sale on Earth. The initiative, called World Product Graph, was revealed on Friday, November 13, in a Reuters exclusive based on an internal company document reviewed by Reuters and accounts from people familiar with the matter.

If fully implemented, the project could turn Amazon into what the document describes as the true “everything store,” cutting into Google’s dominance in online search and fending off Walmart’s growing e-commerce business. It is also aimed squarely at the rise of AI shopping agents, software that can complete shopping tasks on its own for users.

What Is World Product Graph?

World Product Graph “depends on discovering how brands organize their products on their own websites,” according to the document. That would involve some brands and retailers contributing product data to Amazon continuously, but also Amazon independently crawling other retailers’ websites. Reuters could not determine how Amazon planned to crawl the sites or whether it would need to seek permission in all cases.

Amazon has already signed up Coca-Cola and Bayer, among others, to contribute product data, the document says. The company also held a “CEO-level P&G/Amazon summit” for a trial with Procter & Gamble. The companies did not respond to requests for comment.

The project sits in Amazon’s Selection & Catalog Systems unit and is being organized under vice president Robert Tekiela. Amazon is shuffling staff internally to get the project off the ground, according to one of the people familiar with the matter.

Why Amazon Wants the Data

The document argues that Amazon cannot be the go-to destination for every shopping mission while customers must leave the site to research products elsewhere. With more complete data, Amazon could help shoppers determine with more certainty which is the most recent model or colorway of a vacuum cleaner or tablet computer, and guide them toward more subjective choices, such as which product is best for them based on their shopping history.

The numbers explain the urgency. According to a 2025 Morgan Stanley research report cited by Reuters, only 16% of online shoppers begin their product research on Amazon, dwarfed by Google’s 57%. But when customers know exactly what product they plan to buy, Amazon has the advantage, with 41% of searches starting there, compared with 23% for Google and 18% for Walmart.

The project would also address problems with Amazon’s own AI-powered search engine, known as Rufus, which the company began calling “Alexa for Shopping” in May. “Many Rufus answers are factually wrong or incomplete, but customers lack expertise to detect these errors,” the document says. Users conduct about 83 million searches weekly using Rufus, and running them is “expensive,” according to the document. Amazon pushed back, saying “the usage figures are significantly outdated” and that its data shows Alexa for Shopping is roughly 97% accurate globally. Chief executive Andy Jassy has said the assistant has been used by more than 400 million people in the past year.

Battling the AI Agents

World Product Graph would also help stave off the growing threat of alternative AI shopping agents. Jassy sought during the company’s third-quarter earnings call to downplay the impact of agents, saying “few customers are doing much shopping this way yet” and that Amazon’s advertising business has been minimally affected.

Yet Amazon executives have grown wary of AI agents that may send customer traffic to other sites with more complete data, according to people familiar with the matter. Meta’s Muse agent has proven popular, garnering more than 10 million users in the United States and Canada since launching in September. Amazon blocked Muse from its retail site that same month and last year sued AI firm Perplexity to prevent similar access, a case that is still ongoing. JPMorgan said in a research note that Muse “has the potential to become the most widely used consumer AI application since ChatGPT.” Responding to a post saying xAI’s Grok agent was blocked on Amazon, Elon Musk wrote on X last month: “Maybe we should start a competitor to Amazon.”

By creating the world’s most extensive storehouse of product data, Amazon could ensure AI agents working on behalf of shoppers will be compelled to seek products out on Amazon.com and not other retailers’ sites, the people said. That would also help protect Amazon’s growing advertising business.

How Real Is the Plan?

The document is what Amazon calls a PR/FAQ, a six-page narrative memo laying out the rationale and business case for a new product or service. Not all PR/FAQs lead to a service being introduced, and the people familiar with the matter said World Product Graph could be modified or aborted due to financial or other concerns. Reuters could not determine if or when the plan may be implemented, though the document, written like a press release, has an aspirational future release date of October 1, 2027.

Amazon, in a statement, said: “Draft, outdated documents often paint an incomplete picture and the existence of a brainstorming document such as this one doesn’t necessarily reflect our current plans.”

Employees have described World Product Graph as the beginnings of an effort to make the site a one-stop shop for buying anything imaginable, an expansion of existing services known as “Buy for Me” and “Shop Direct,” which list products sold on other sites on Amazon, allowing customers to buy them with Amazon as an intermediary. “Customers have always been able to find products on Amazon,” Tekiela said in the document. “What’s been missing is understanding — how do these products relate to each other, which one is right for me, and what else should I consider?”

The push comes as the biggest tech companies race to build out AI infrastructure — SpaceX alone is planning more than $74 billion in AI supercomputing campuses in Texas and Arizona. It also follows a week of high-profile technology disruptions, including the Microsoft 365 outage that knocked out Outlook and Teams for millions of users.

Why This Matters

For shoppers, a fully built World Product Graph could mean richer product comparisons and fewer wrong answers from AI assistants, all without leaving Amazon. For Google and Walmart, it would mean a direct challenge to their role as the starting point for product research. And for the fast-growing field of AI shopping agents, it could decide whether the agents of the future shop on Amazon’s terms or someone else’s.

But the plan is not a product yet. Amazon itself says the document is a draft that may not reflect its current plans, and Reuters could not confirm a launch date. What is confirmed is the direction of travel: Amazon sees control of product information as the next battleground in online shopping, and it is moving staff and signing up brands to fight for it.

Frequently Asked Questions

What is Amazon’s World Product Graph?

World Product Graph is an internal Amazon initiative to collect product data on virtually every item for sale worldwide, whether or not Amazon sells it, according to an internal document reviewed by Reuters. The goal is to make Amazon the starting point for all product research and shopping, and to counter AI shopping agents that could send customers elsewhere.

When will World Product Graph launch?

The internal document carries an aspirational release date of October 1, 2027, but Reuters could not determine if or when the plan will be implemented. Amazon said the document is a draft that “doesn’t necessarily reflect our current plans,” and people familiar with the matter said the project could be modified or abandoned.

Why is Amazon worried about AI shopping agents?

AI agents can complete shopping tasks on their own for users, and they could direct shoppers to whichever site has the most complete product data. Meta’s Muse agent has already passed 10 million users in the US and Canada since September, and Amazon blocked Muse from its retail site and sued Perplexity over similar access.

Sources: Reuters (via syndication); Global Banking and Finance; The Outpost; Traders Union.

Leave a Reply

Your email address will not be published. Required fields are marked *