Generative AI Digital Assistant Function Dependency Handling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing digital assistants face limitations in handling diverse user inputs and complex business scenarios due to reliance on explicit conversation designs, leading to inefficiencies, errors, and increased operational costs, particularly when dealing with interdependent functions and sensitive business terminology.

Innovation Solution

A digital assistant system leveraging a generative AI model to dynamically interpret user inputs and manage function dependencies, separating context into model-accessible and non-model-accessible data to reduce context size and ensure accurate, efficient function execution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If explicit conversation design is used to handle user inputs, then the digital assistant can follow predetermined flows, but it becomes unable to correctly map diverse user inputs to intended actions and integrate contextual data

Engineering Contradiction:

Improveaccuracy of mapping user input to actionVSAvoidability to handle diverse user inputs

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent replaces the mechanical explicit conversation design system with a generative AI model that uses natural language processing to understand and respond to user inputs. The generative AI model processes user messages, determines intents, and generates responses without being constrained by predetermined conversation flows, thereby achieving both accuracy and adaptability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the operational parameters of the digital assistant by transitioning from a rigid flow-based architecture to a generative AI-based architecture. This parameter change enables the system to dynamically adapt to diverse user inputs while maintaining accurate intent recognition and action mapping through the AI model's learning capabilities.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If all context data is provided to the generative AI model, then the model can make accurate decisions, but the context size becomes large leading to increased costs and slower response times

Engineering Contradiction:

Improveaccuracy of model decisionsVSAvoidresponse time and operational cost

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system extracts and separates critical context data from the full conversation history, providing only the most relevant information to the generative AI model. This extraction process reduces context size while maintaining decision accuracy by focusing on essential data points needed for the current user input.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The context data is segmented into different components, with only the necessary portions being provided to the generative AI model. This segmentation allows the system to manage context efficiently by dividing the information into relevant and irrelevant parts, thereby reducing processing overhead and costs.

Inventive Principle:

Principle #1Segmentation

3Ease of operation

If the digital assistant uses a predetermined conversation flow, then it can handle structured interactions, but it becomes unable to recognize dependencies between data items and integrate contextual data across messages

Engineering Contradiction:

Improveability to handle structured interactionsVSAvoidability to recognize dependencies and integrate context

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system transitions from a static predetermined conversation flow to a dynamic generative AI-based interaction model. This dynamic approach allows the digital assistant to adapt its responses based on the current context, user input, and identified dependencies between data items, enabling both structured handling and flexible integration of contextual information.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The generative AI model acts as an intermediary between the user input and the digital assistant's response generation. It processes the user message, integrates contextual data across messages, recognizes dependencies between data items, and generates appropriate responses, thereby bridging the gap between structured interaction handling and adaptive context integration.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250321768A1Digital assistant service using generative artificial intelligence

Publication Date: 2025.10.16 SAP SE

AI summary

Examples described herein relate to a digital assistant that utilizes generative artificial intelligence. Prompt data provided to a generative machine learning model includes user input and function data. The function data can include dependency data that identifies at least one function dependency. A first function is invoked based on a first response from the generative machine learning model to obtain first output data. After updating the prompt data to include the first output data and receiving a second response from the generative machine learning model, a second function is invoked to obtain second output data. The digital assistant can maintain model-accessible data and non-model-accessible data for a digital conversation. Automated validation can be performed on parameter values of the first function or the second function. Parameter values may be explicitly confirmed by the digital assistant via a user-confirmation operation before invoking the first function or the second function.