AI Assistant Service Configuration Using Master Program Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing AI assistant configurations require advanced programming skills and are inefficient for building services that involve multiple language-based tasks, necessitating frequent reconfiguration for each new service or procedure.

Innovation Solution

A method and apparatus that utilize a generic master program and separate procedural knowledge information to configure AI assistants, allowing service creation without programming skills, by using a structured descriptive language for task sequences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional AI assistant configuration methods are used, then the AI assistant can perform specific tasks, but advanced programming skills are required and reconfiguration is needed for each new service

Engineering Contradiction:
Improveability to perform different servicesVSAvoidconfiguration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the AI assistant configuration into two independent parts: a master program containing generic task-execution logic and procedural knowledge containing service-specific instructions. This segmentation allows the procedural knowledge to be modified for different services without changing the master program, reducing configuration complexity while maintaining versatility.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The master program is designed as a universal component that can execute any service by loading different procedural knowledge. The same master program structure handles diverse services (information retrieval, data analysis, content creation) by simply changing the procedural knowledge input, eliminating the need for reconfiguration of the core execution logic.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Reliability

If custom configurations are created for each service, then specific services can be performed accurately, but the process becomes time-consuming and inefficient

Engineering Contradiction:
Improveservice execution accuracyVSAvoidconfiguration time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by pre-structuring the master program with all necessary execution logic, task templates, and decision-making frameworks before service deployment. This pre-prepared structure enables rapid configuration of new services by simply filling in procedural knowledge without time-consuming setup, while maintaining execution accuracy through the validated master program framework.

Inventive Principle:
Principle #10Preliminary action

3Ease of manufacture

If a generic master program is used, then reconfiguration for new services is simplified, but the program must be adaptable to various task types

Engineering Contradiction:
Improveservice deployment easeVSAvoidtask type flexibility
Core Design Contradiction:
Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The master program incorporates dynamic elements including adaptable task templates, flexible decision-making logic, and parameterized task execution frameworks. These dynamic components allow the generic master program to automatically adjust to different task types by loading appropriate procedural knowledge, maintaining both ease of deployment and task flexibility.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4650955A1Method for configuring a service to be delivered to a user and artificial intelligence assistant for delivering the same
Publication Date: 2025.11.19 OVERSTAND K
  • EP4650955A1 patent drawingFigure 1
  • EP4650955A1 patent drawingFigure 2
  • EP4650955A1 patent drawingFigure 3

AI summary

Computer-implemented method for providing configuration information (MP, PK) related to a service to be delivered to a user by an Artificial Intelligence, AI, assistant being configured to provide a response to a user request, or prompt, received through a user interface of the AI assistant to at least said user, by applying said user request to a trained Language Model, LM, said method comprising the steps of: - obtaining (24) first instructions (MI) to get the AI assistant (AS) retrieve a master program (MP), said master program comprising pseudo-code instructions to be charged into the AI assistant (AS), said pseudo-code instructions being generic to any service to be delivered to a user and configured to cause the AI assistant (AS) to run the sequence of tasks defined in procedural knowledge information (PK), external to the master program, - obtaining (25) second instructions (PKI) to get the AI assistant retrieve said knowledge procedural information (PK), separately from said master program (MP), said knowledge procedural information being specific to the service and comprising a structured description of a sequence of tasks defining said service, and - providing (26) said first and second instructions (MI, PKI) for transmission to said AI assistant through a communication interface.