Distributed Ledger AI Risk Classification Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing use of artificial intelligence (AI) and machine learning models poses unknown risks that existing information processing systems are unable to effectively address or present to users.
Innovation Solution
An information processing system utilizing a distributed network with a distributed ledger that stores transaction data regarding attribute information related to AI risk classifications, allowing for the display of risk information associated with learned models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a distributed ledger is used to store AI risk classification data, then reliability and tamper-proof recording are improved, but device complexity increases
Solution Approach 1:
The system segments the complex distributed ledger functionality into a specialized AI risk classification management module. This module handles only the specific task of storing and managing AI risk classification data, separating it from other system functions. The segmentation allows the system to leverage the reliability of distributed ledger technology for a specific purpose without requiring the entire system to be rebuilt as a complex distributed ledger, thus improving reliability while controlling overall system complexity.
Solution Approach 2:
The patent introduces an intermediary component that acts as a bridge between the distributed ledger and the AI risk classification system. This intermediary handles the complexity of distributed ledger operations (such as transaction validation, block creation, and consensus mechanisms) and presents a simplified interface to the risk classification module. By placing this intermediary layer, the system achieves reliable tamper-proof recording through the distributed ledger while shielding the rest of the system from its inherent complexity.
2Loss of information
If distributed ledger technology is implemented for AI risk tracking, then information transparency is improved, but loss of time in data management increases
Solution Approach 1:
The system performs preliminary actions by pre-defining risk classification categories and criteria before actual AI risk data needs to be recorded. The distributed ledger is pre-configured with the structure and validation rules for AI risk classifications. When risk events occur, they can be directly mapped to pre-established categories and recorded immediately without requiring complex real-time analysis or classification decisions, thus improving information transparency while minimizing data management time.
Solution Approach 2:
The patent utilizes parameter changes by transforming complex AI risk assessment data into standardized, discrete classification parameters that are suitable for efficient storage in the distributed ledger. Instead of storing raw, unstructured risk assessment data, the system converts it into standardized risk levels, categories, and indicators that can be quickly processed and recorded. This parameter transformation enables rapid data management while maintaining full transparency of the risk information in the distributed ledger.
Data Source
AI summary
The disclosure relates to an information processing system. The information processing system includes a distributed network that implements a distributed ledger. The distributed ledger stores one piece of transaction data regarding attribute information that indicates one classification related to one learned model out of a plurality of classifications regarding risks posed by AI.