USPTO Patent Granted for Network Content Generation Methods

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 25th, 2026

Summary

The USPTO has granted a patent to Stripe, LLC for systems and methods facilitating network content generation using a dynamic multi-model approach. This patent covers the use of machine learning and heuristic models to predict user behavior and generate personalized content.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has granted patent US12586093B2 to Stripe, LLC, covering "Systems and methods for facilitating network content generation via a dynamic multi-model approach." The patent details a multi-model system, potentially incorporating machine learning and heuristic models, designed to generate content for users with greater accuracy. The abstract describes a process where user data is analyzed by a selected model to produce a predicted result, which then informs the generation of content displayed on a user's device.

This patent grant signifies a new intellectual property asset for Stripe, LLC in the domain of AI-driven content generation. While patent grants themselves do not impose new regulatory obligations on other entities, they can influence industry practices and competitive landscapes. Companies operating in AI and content generation may wish to review the patent's claims to understand the scope of protection and potential implications for their own product development and intellectual property strategies.

Source document (simplified)

← USPTO Patent Grants

Systems and methods for facilitating network content generation via a dynamic multi-model approach

Grant US12586093B2 Kind: B2 Mar 24, 2026

Assignee

Stripe, LLC

Inventors

Mengjie Ding, Bo Xu, Jack Dent

Abstract

In some embodiments, a multi-model approach may be utilized to produce predictions with greater accuracy, which may then be used to generate content for one or more entities. In some embodiments, such models may include one or more machine learning models, heuristic models, exponential smoothing models, and/or other models. As an example, based on a processing volume corresponding to user data associated with a user, a first model may be selected for processing the user data over use of a second model for processing the user data. As such, based on the selection of the first machine learning model, the user data may be inputted into the first machine learning model to obtain a predicted result related to the user. Based on the predicted result, content for the user may be generated for display on a user device associated with the user.