Natural-Language Contract Execution Using LLM Event Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing insurance business using decentralized finance (DeFi) faces inefficiencies in designing and developing smart contracts due to the need for contracts to be written in specialized languages like Solidity, lacking mechanisms for automatically executing contracts based on natural language.

Innovation Solution

A contract management system utilizing a contract execution support device with a large language model to process natural language inputs, determine contract events, and generate necessary execution information, integrated with a contract execution system for consensus-based execution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If contracts are written in specialized languages like Solidity for smart contract execution, then contract execution reliability is improved, but design and development time increases

Engineering Contradiction:

Improvecontract execution reliabilityVSAvoiddesign and development time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent introduces a large language model as an intermediary between natural language contract documents and specialized smart contract code. The LLM translates natural language contract terms into executable smart contract code, eliminating the need for users to directly write specialized languages while maintaining execution reliability through automated code generation from verified natural language specifications.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical process of manually writing and verifying smart contract code in specialized languages with an automated AI-based translation system. The large language model substitutes for human developers in the code generation task, transforming the manual mechanical process into an automated intelligent system that reduces time consumption while preserving contractual execution reliability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If contracts are written in natural language for ease of understanding, then ease of operation is improved, but automation of contract execution decreases

Engineering Contradiction:

Improveease of contract understandingVSAvoidcontract execution automation

Core Design Contradiction:

Ease of operationVSExtent of automation

Solution Approach 1:

The large language model serves as a bridge intermediary that accepts natural language contract documents as input and automatically generates executable smart contract code. This intermediary system enables contracts written in human-readable natural language to be automatically translated into machine-executable code, thus maintaining ease of operation while achieving full automation of contract execution through the LLM's code generation capabilities.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If manual review of contract execution information is performed, then accuracy of contract execution is improved, but processing time increases

Engineering Contradiction:

Improveaccuracy of contract executionVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system implements self-service by having the large language model automatically extract, verify, and prepare contract execution information without requiring manual review. The LLM independently processes contract documents, identifies execution conditions, and generates execution instructions, eliminating the need for human intervention in the review process while maintaining high accuracy through the model's intelligent processing capabilities.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical review process with an automated AI-based information extraction and verification system. The large language model substitutes for human reviewers in extracting contract execution information, verifying contract conditions, and preparing execution data, thereby reducing processing time while maintaining accuracy through intelligent automated processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250378269A1Contract managment system and contract management method

Publication Date: 2025.12.11 HITACHI LTD

AI summary

A contract management system acquires application information, which is information obtained from an applicant applying for the execution of a contract, written in natural language, and extracts contract event information, which is information related to a contract event, which is an event that causes the execution of the contract, from the contract written in natural language. The contract management system inputs the application information and the contract event information extracted from the contract into a large language model to determine whether the application information includes a contract event. If the application information includes a contract event, the contract management system acquires contract execution information, which is information necessary for the execution of the contract corresponding to the contract event, by inputting the application information into the large language model, and inputs the acquired contract execution information into the contract execution system.