Generative User Routines With Context-Aware Device Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing generative models lack personalization and tailoring of generative output, especially when users seek highly personalized content, due to the use of generic external tools that do not account for individual user contexts or device capabilities.

Innovation Solution

A generative model is utilized to generate, update, and execute user routines based on user input, device metadata, and application capabilities, selecting devices and applications to perform actions that align with the user's desired goals, even if not explicitly mentioned, and adjusting these routines in response to changes in user inputs, environment, or deviations from the routine.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If generic external tools are used to expand context beyond explicit user input, then the system can obtain additional content (e.g., search results) to improve generative output quality, but the generative output becomes impersonalized and not tailored to individual user contexts or device capabilities

Engineering Contradiction:

Improvequality of generative outputVSAvoidpersonalization to user context

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system segments the context into multiple layers: explicit user input, implicit user context (preferences, history, device metadata), and external tool results. Each layer is processed separately and integrated by the generative model to produce personalized output.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by proactively gathering implicit user context and device metadata before generating output. This includes retrieving user preferences, interaction history, and device capabilities in advance, so the generative model has personalized context ready when processing user input.

Inventive Principle:

Principle #10Preliminary action

2Extent of automation

If the system proactively identifies and selects devices and applications to perform user routine actions without explicit user specification, then the level of automation and personalization increases, but the system complexity and computational resources required for context processing increase

Engineering Contradiction:

Improveautomatic device selectionVSAvoidsystem complexity

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The generative model serves multiple functions: it processes user input, infers implicit context, selects appropriate devices and applications, and generates actionable routines. This multi-functionality reduces the need for separate specialized modules, managing system complexity while maintaining high automation.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system performs self-service by automatically analyzing user input and context to identify appropriate devices and applications without requiring explicit user specification. The generative model autonomously makes decisions about device selection and routine configuration based on available context.

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If the system dynamically updates user routines based on additional user input and environmental changes, then the adaptability and user experience improve, but the computational resources and processing time required increase

Engineering Contradiction:

Improvedynamic routine adjustmentVSAvoidcomputational resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The system implements periodic action by updating routines at appropriate intervals and triggers rather than continuously. Routines are updated when new user input is received, when environmental changes are detected, or when scheduled review periods occur, reducing unnecessary computational processing while maintaining adaptability.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system uses feedback mechanisms to monitor user interactions and routine execution outcomes. This feedback informs incremental updates to routines, allowing the system to adapt to user needs while optimizing resource usage by making targeted adjustments rather than complete re-processing of all context.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20260010394A1System(s) and method(s) for utilization of generative model(s) in generating, updating, and/or executing user routine(s)

Publication Date: 2026.01.08 GOOGLE LLC

AI summary

Implementations relate to receiving user input from a user that describes at least one type of action to be routinely performed, but without identifying any device or application in association with the at least one action, and in response, utilizing generative model(s) to determine action(s) to be performed by device(s) and/or applications, that are associated with the user, and in furtherance of executing a user routine. The action(s) can be determined based on processing, using the generative model(s), the user input and metadata associated with device(s) and/or application(s) that indicates capabilities of the device(s) and/or application(s). The user routine can be periodically modified or updated based on additional user input(s) and/or based on monitored performance (or lack thereof) of the user routine.