/Patent Tracker
AmazonGranted patentAI & softwareInterest 7/10

Amazon patents shopping assistant that remembers your search filters

This patent describes a system that tracks your actions on a search results page, such as clicking a specific item or applying a filter. It feeds this real-time information into a generative model to create suggested questions or comparisons, which appear in a chat-like box on the screen.

Why it mattersThis could make online shopping feel more like a conversation, helping users find the right product without typing new searches for every change in preference.

In the filing’s own words

Systems and methods are described herein for implementing context-aware generative model ingress operations. During a respective browsing session, a context data object comprising contextual data related to the browsing session may be updated in real-time based on various interactions with a web application, and the contextual data may be provided to a generative model as part of an input prompt. Ingress points may be placed in various locations within an interface of the web application and interactions with the ingress points may be effective to update the context data object based on various context filtering operations and the respective context associated with the ingress points. Updated contextual data may be passed to the generative model for use in generating contextual search result insights, comparisons, and/or recommendations. Such model output may be displayed via a conversation layer object displayed via the web application.

What exactly they claim (claim 1)

1. A computer-implemented method comprising: receiving, during a first browsing session, a first search query from a first computing device; causing display of a first search results page (SRP) on the first computing device, wherein the first SRP comprises a plurality of search results identified based at least in part on the first search query; storing first session context data in a context data object associated with the first browsing session, wherein the first session context data comprises at least first SRP data representative of the plurality of search results; detecting a first selection event associated with a first interactive display element displayed on the first SRP, wherein the first interactive display element is associated with a first search result of the plurality of search results, and wherein the first selection event is associated with a first filtering operation; storing, based at least in part on the first filtering operation, second session context data in the context data object, wherein the second session context data comprises at least first detail data representative of the first search result; providing a first input prompt comprising at least the second session context data to a generative model; storing at least first suggested query response data associated with a first suggested query in a memory cache, wherein the first suggested query response data and the first suggested query are generated by the generative model based on the first input prompt; and causing display of a conversation layer object on the first SRP, wherein the conversation layer object comprises at least a second interactive display element representing the first suggested query and at least a third interactive display element representing the first search result.

Facts

Company
Amazon (Amazon Technologies, Inc.)
Official title
Context-aware generative model ingress system
Patent number
US12754966B1
Application number
19300163
Filed
Aug 14, 2025
Granted
Oct 6, 2026
Inventors
6
Claims
20

Classified under

G06F 16/9538 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Amazon in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.