Dynamic experience curation platform

Patent metadata

Field	Value
Publication number	US-12468770-B2
Application number	US-202418636264-A
Country	US
Kind code	B2
Filing date	Apr 16, 2024
Priority date	Feb 15, 2024
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

A dynamic experience curation platform utilizing AI and ML techniques, including neural networks and generative models, to enhance user interactions with content on the Internet. The platform employs an Experience Broker (EB) service to disintermediate user devices from the internet (or other content/information sources), improving security and user experience. It processes content requests, extracts relevant information, filters out unwanted content like ads, transforms, combines, and/or mutates existing content, and generates curated content consistent with user preferences and other constraints. The platform's generative AI process renders content into a user interface consistent with user preferences. It allows for personalized content delivery across devices, including virtual and augmented reality environments. The platform supports a spectrum of personalization, allowing users to fine-tune their content consumption experience. Additionally, it manages user sessions across devices and integrates various databases for storing user profiles, preferences, and other relevant information.

First claim

What is claimed is:

1 . A computing system for dynamic experience curation using one or more virtual cross-platform agents employing a dynamic experience curation platform, the computing system comprising:
one or more hardware processors configured for:
transmitting a content request to one or more content providers;
receiving the requested content from the one or more content providers;
feeding the requested content or a subset thereof into a classification model to extract and classify relevant content;
retrieving a plurality of user preference data;
verifying an applicable preference hierarchy from a plurality of stored preference sets to determine a contextually relevant preference set active for the current content request based on current context, wherein current context comprises a sliding time window that includes an amount of historical, current, and projected context based on a finite time horizon;
engineering at least one prompt or input using an output of the classification task and the plurality of resultant user preference data;
submitting the prompt and the requested content to one or more generative artificial intelligence (AI) models;
receiving output from the one or more generative AI models; and
rendering, on a user interface of a user device, the received output as curated content responsive to the content request.

2 . The computing system of claim 1 , wherein the user device is a first user device and the content request is received from the first user device.

3 . The computing system of claim 2 , wherein the output of the one or more generative AI models is rendered on a second user device.

4 . The computing system of claim 1 , wherein the one or more generative AI models include a Retrieval Augmentation Generation component.

5 . The computing system of claim 1 , wherein the rendered output is rendered directly or indirectly via one or more existing applications.

6 . The computing system of claim 1 , wherein the prompt comprises text, a neural network interface to an AI system, a binary object, or a machine understandable image or video format.

7 . The computing system of claim 1 , wherein the curated content comprises accessibility overlays.

8 . The computing system of claim 7 , wherein the accessibility overlays comprise text-to-speech conversion, a screen reader, alternative control/input methods, magnification and zoom, color contrast adjustment, captions and transcripts, sign language interpretation, keyboard navigation, voice control, simplified content layout, readability enhancements, visual cues and notifications, haptic feedback, and easy language and symbol support.

9 . The computing system of claim 1 , wherein the one or more virtual cross-platform agents implement application-to-application communication protocols for coordinating with external systems.

10 . The computing system of claim 1 , wherein the dynamic experience curation platform provides a natural language interface for web resource interaction.

11 . The computing system of claim 1 , wherein the system implements model context protocols for managing communication between the generative AI models and external data sources.

12 . A computer-implemented method executed on a dynamic experience curation platform for dynamic experience curation using one or more virtual cross-platform agents, the computer-implemented method comprising:
transmitting a content request to one or more content providers; receiving the requested content from the one or more content providers; feeding the requested content or a subset thereof into a classification model to extract and classify relevant content; retrieving a plurality of user preference data; engineering a prompt using an output of the classification task and the plurality of user preference data; verifying an applicable preference hierarchy from a plurality of stored preference sets to determine a contextually relevant preference set active for the current content request based on current context, wherein current context comprises a sliding time window that includes an amount of historical, current, and projected context based on a finite time horizon; submitting the prompt and the requested content to one or more generative artificial intelligence (AI) models; receiving output from the one or more generative AI models; and rendering, on a user interface of a user device, the received output as curated content responsive to the content request.

13 . The computer-implemented method of claim 12 , wherein the user device is a first user device and the content request is received from the first user device.

14 . The computer-implemented method of claim 13 , wherein the output of the one or more generative AI models is rendered on a second user device.

15 . The computer-implemented method of claim 12 , wherein the one or more generative AI models include a Retrieval Augmentation Generation component.

16 . The computer-implemented method of claim 12 , wherein the rendered output is rendered directly or indirectly via one or more existing applications.

17 . The computer-implemented method of claim 12 , wherein the prompt comprises text, a neural network interface to an AI system, a binary object, or a machine understandable image or video format.

18 . The computer-implemented method of claim 12 , wherein the curated content comprises accessibility overlays.

19 . The computer-implemented method of claim 18 , wherein the accessibility overlays comprise text-to-speech conversion, a screen reader, alternative control/input methods, magnification and zoom, color contrast adjustment, captions and transcripts, sign language interpretation, keyboard navigation, voice control, simplified content layout, readability enhancements, visual cues and notifications, haptic feedback, and easy language and symbol support.

20 . A system for dynamic experience curation using one or more virtual cross-platform agents employing a dynamic experience curation platform, comprising one or more computers with executable instructions that, when executed, cause the system to:
transmit a content request to one or more content providers; receive the requested content from the one or more content providers; feed the requested content or a subset thereof into a classification model to extract and classify relevant content; retrieve a plurality of user preference data; engineer a prompt using an output of the classification task and the plurality of user preference data; verify an applicable preference hierarchy from a plurality of stored preference sets to determine a contextually relevant preference set active for the current content request based on current context, wherein current context comprises a sliding time window that includes an amount of historical, current, and projected context based on a finite time horizon; submit the prompt and the requested content to one or more generative artificial intelligence (AI) models; receive output from the one or more generative AI models; and render, on a user interface of a user device, the received output as curated content responsive to the content request.

21 . The system of claim 20 , wherein the user device is a first user device and the content request is received from the first user device.

22 . The system of claim 21 , wherein the output of the one or more generative AI models is rendered on a second user device.

23 . The system of

Assignees

• Qomplx Llc

Inventors

• Crabtree Jason
• Kelley Richard
• Hopper Jason
• Park David

Classifications

• G09B21/00

  Teaching, or communicating with, the blind, deaf or mute (audible presentation of material to be studied G09B5/04; devices or methods for replacing direct visual or auditory perception by another kind of perception A61F9/08, A61F11/04; audible indication of meter readings or of colour G01D7/12; watches for blind persons G04B25/02; speech analysis, speech recognition G10L; sound-recording or reproducing, per se G11B) · CPC title

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title