Amazon aims to build global database of all online products

Amazon plans to build a massive vault of data on every product sold anywhere on the internet, even items the company does not sell itself. This project is called the World Product Graph. The goal is to make the information available on Amazon so shoppers stay on the site instead of doing product research on Google or Reddit. If people search for an item sold elsewhere, the company still hopes they will end up on its pages.

AI Agents

This strategy also targets AI agents. If Amazon holds the biggest stockpile of shopping data, bots may rely on its product index over others. Users shopping with AI assistants will still end up on Amazon pages. Outside retailer links already appear on the site through a shop direct feature launched last year. Amazon also runs an AI shopping assistant, Alexa for Shopping, which was previously called Rufus. An internal document notes that many Rufus answers are wrong or incomplete, and running user searches with the AI is costly. The new project aims to fix these issues by giving the system more data to improve accuracy and allow cheaper models.

Launch Plans

Whether this project actually launches remains uncertain. An internal document states that the plan depends on figuring out how brands organize products on their own websites, which could mean crawling other rival sites. Reuters reports that it is unclear if or when the plan will happen, though an internal document listed an October 1st, 2027 release date. Amazon shared a statement saying that draft and outdated documents often show an incomplete picture, and a brainstorming document does not necessarily reflect current company plans.

Context

Large technology companies constantly work to gather more data to keep users inside their own digital ecosystems for as long as possible. Online product search has shifted away from traditional search engines as more users rely on social platforms and artificial intelligence tools for recommendations. Access to complete and accurate information across the entire web is a key challenge for companies building automated shopping assistants.