Automated Assistant Consent Routing for Secure Cloud-Local Fulfillment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automated assistants face challenges in dynamically switching between cloud-based and local execution to balance data security and optimal content provision, leading to potential data security risks and sub-optimal responses.

Innovation Solution

Implementing a secure automated assistant that prompts users for consent before transmitting data to servers or third-party applications based on the category of the assistant command, using machine learning models to determine the category and rules to manage data transmission, ensuring secure and efficient data handling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If automated assistants transmit data to cloud servers for processing, then response quality and content accuracy are improved, but data security and user privacy are compromised

Engineering Contradiction:
Improveresponse qualityVSAvoiddata security risk
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The system dynamically changes the execution location parameter between cloud and device based on data sensitivity classification. By categorizing data into different sensitivity levels, the system adjusts whether processing occurs in the cloud or locally, optimizing both response quality and security for each data type.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The automated assistant system is segmented into cloud-based and device-based components. Sensitive data processing is separated from non-sensitive data processing, allowing the system to leverage cloud resources for general tasks while keeping sensitive operations local to maintain security.

Inventive Principle:
Principle #1Segmentation

2Object-affected harmful factors

If automated assistants execute exclusively locally on client devices, then data security is preserved, but response quality and content availability become sub-optimal

Engineering Contradiction:
Improvedata securityVSAvoidresponse quality
Core Design Contradiction:
Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The system dynamically changes the execution location parameter between cloud and device based on data sensitivity classification. By categorizing data into different sensitivity levels, the system adjusts whether processing occurs in the cloud or locally, optimizing both response quality and security for each data type.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system transitions from a static execution model (either exclusively cloud or exclusively device) to a dynamic model that adapts execution location based on real-time data sensitivity assessment. This allows the system to optimize security and response quality on a per-task basis.

Inventive Principle:
Principle #15Dynamics

3Device complexity

If automated assistants cannot dynamically switch between cloud and local execution, then system complexity is reduced, but adaptability to different security and performance requirements is limited

Engineering Contradiction:
Improvesystem complexityVSAvoiddynamic switching capability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The system dynamically changes the execution location parameter between cloud and device based on data sensitivity classification. By categorizing data into different sensitivity levels, the system adjusts whether processing occurs in the cloud or locally, optimizing both response quality and security for each data type.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system transitions from a static execution model (either exclusively cloud or exclusively device) to a dynamic model that adapts execution location based on real-time data sensitivity assessment. This allows the system to optimize security and response quality on a per-task basis.

Inventive Principle:
Principle #15Dynamics

4Measurement precision

If automated assistants transmit user data to servers, then optimal content can be provided, but user consent and control over data transmission are reduced

Engineering Contradiction:
Improvecontent accuracyVSAvoiduser control
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system implements automatic data sensitivity classification and execution location selection, eliminating the need for users to manually navigate settings or make complex decisions about data transmission. The automated classification system handles security decisions in the background while maintaining optimal performance.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4143823B1Methods and systems for providing a secure automated assistant
Publication Date: 2026.03.25 GOOGLE LLC
  • EP4143823B1 patent drawingFigure 1
  • EP4143823B1 patent drawingFigure 2
  • EP4143823B1 patent drawingFigure 3A

AI summary

Implementations described herein relate to receiving user input directed to an automated assistant, processing the user input to determine whether data from a server and/or third-party application is needed to perform certain fulfillment of an assistant command included in the user input, and generating a prompt that requests a user consent to transmitting of a request to the server and/or the third-party application to obtain the data needed to perform the certain fulfillment. In implementations where the user consents, the data can be obtained and utilized to perform the certain fulfillment. In implementations where the user does not consent, client data can be generated locally at a client device and utilized to perform alternate fulfillment of the assistant command. In various implementations, the request transmitted to the server and/or third-party application can be modified based on ambient noise captured when the user input is received.