Information Provision Device with Segmented AI Parameter Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods to protect learned artificial intelligence from falsification or theft are inadequate, relying primarily on trade secrets and contracts, which are insufficient for robust security.
Innovation Solution
A system utilizing a distributed recording system with a blockchain architecture to securely store and manage parameter groups and design information for artificial intelligence, enabling generation and transmission of support information while encrypting and dividing the intelligence into parts to enhance protection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If learned artificial intelligence is protected only by trade secrets and contracts, then service providers must make continuous efforts to protect the intelligence themselves, but this protection method is insufficient and vulnerable to falsification or theft by third parties
Solution Approach 1:
The learned artificial intelligence is divided into multiple parameter groups that are distributed and recorded in separate recording devices within a distributed recording system. This segmentation ensures that no single device holds the complete intelligence, making theft or falsification more difficult while maintaining system reliability.
Solution Approach 2:
A distributed recording system acts as an intermediary between the service provider and the learned artificial intelligence. This intermediary layer provides robust protection against falsification and theft by third parties, eliminating the need for service providers to implement their own separate protection mechanisms.
2Ease of operation
If the complete learned artificial intelligence is stored in a single location, then access and usage are simplified, but the intelligence becomes vulnerable to falsification and theft
Solution Approach 1:
The learned artificial intelligence is segmented into multiple parameter groups distributed across different recording devices. This segmentation reduces the risk of falsification and theft while maintaining ease of operation through automated reconstruction when needed.
Solution Approach 2:
Different recording devices store different parameter groups with specific local characteristics. Each device holds only a portion of the intelligence, and the system automatically reconstructs the complete intelligence when access is required, balancing security with operational convenience.
3Reliability
If parameter groups are distributed across multiple recording devices, then protection against falsification and theft is enhanced, but the system complexity increases
Solution Approach 1:
The distributed recording system serves multiple functions: it stores parameter groups, protects against falsification and theft, and automatically reconstructs the complete learned artificial intelligence when needed. This multi-functionality reduces the need for separate protection systems, thereby managing complexity while enhancing reliability.
4Productivity
If all parameter groups are made accessible for generating support information, then service quality is improved, but the risk of unauthorized access and data breach increases
Solution Approach 1:
Parameter groups are segmented and distributed across multiple recording devices, allowing the system to reconstruct only the necessary portions for generating support information. This segmentation enables efficient service provision while minimizing the exposure of complete intelligence data, thereby reducing unauthorized access risk.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is an information provision device, comprising: a parameter group acquisition unit for referencing one storage device included in a distributed storage system wherein parameter groups that characterize a learned artificial intelligence is stored in advance in a plurality of storage devices in a distributed manner, and acquiring the relevant parameter group; a design information acquisition unit for acquiring design information wherein is defined information capable of identifying the applicable area in the learned artificial intelligence for each of the parameter groups; an artificial intelligence building unit for building, on the basis of the acquired parameter group and design information, at least a portion of the learned artificial intelligence; and an auxiliary information transmission unit for transmitting, to a customer's information processing device, auxiliary information outputted from at least a portion of the built artificial intelligence.