AI Learning System with Blockchain-Encrypted Personal Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI learning systems face challenges in performing AI learning with personal information while complying with legal requirements such as the Personal Information Protection Law and GDPR, as it is difficult to maintain confidentiality and prevent personal information from being identified or leaked.
Innovation Solution
An AI learning system that includes a sorter to identify personal information to be kept secret, an encryptor to encrypt this information, and a holder to store it in a blockchain, with an AI learner performing secure calculations on the encrypted data to create an AI model, all while optimizing calculation processing based on the degree of secrecy and encryption impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If personal information is encrypted to comply with legal requirements, then confidentiality and protection of personal information are improved, but calculation processing complexity and computational cost increase
Solution Approach 1:
The patent segments the AI learning data into two distinct parts: encrypted personal information and non-personal information. This segmentation allows the system to apply encryption only where necessary (to personal information) while leaving other data unencrypted and easily processable, thus reducing overall computational complexity while maintaining confidentiality.
Solution Approach 2:
The patent extracts and separates personal information from the broader AI learning data set. By identifying and extracting only the personal information components, the system can encrypt only these specific portions rather than the entire data set, thereby reducing the computational burden while still achieving the required level of protection.
2Reliability
If encryption is applied to personal information, then legal compliance is improved, but the amount of calculation processing increases
Solution Approach 1:
The patent applies local quality by implementing encryption only on the specific portions of data that require protection (personal information) rather than uniformly encrypting the entire data set. This selective approach maintains legal compliance for personal information while preserving computational efficiency for non-personal data processing.
Solution Approach 2:
The patent employs partial action by encrypting only the necessary portion of data (personal information) rather than the entire data set. This partial encryption approach provides sufficient legal compliance while avoiding the excessive computational cost that would result from encrypting all data regardless of sensitivity.
3Reliability
If AI learning is performed on encrypted data, then data security is improved, but learning accuracy and model performance may deteriorate
Solution Approach 1:
The patent segments the data into encrypted and unencrypted portions, allowing the AI learning model to process non-personal information with full accuracy while maintaining security for personal information. This segmentation enables the model to achieve sufficient learning accuracy without the performance penalties that would result from encrypting the entire data set.
Solution Approach 2:
The patent introduces an intermediary approach where the system processes encrypted and unencrypted data through different computational paths that converge in the learning model. This intermediary handling allows the model to leverage the security benefits of encryption for personal information while maintaining the computational efficiency and accuracy needed for effective learning.
Data Source
AI summary
An AI learning system comprising: a sorter configured to sort, from among AI learning data, personal information that should be secret when creating a desired AI model, according to prescribed criteria; an encryptor configured to encrypts the sorted personal information;a holder configured to hold the AI learning data including the encrypted personal information in a blockchain; and an AI learner configured to perform secret calculations on at least the data portion related to encrypted the personal information to learn the AI model, using the stored AI learning data; wherein the sorting unit sets the prescribed criteria so as to reduce amount of calculation processing, depending on the degree of increase or decrease in the amount of calculation processing of the secure computation due to encryption of the personal information and the degree to which the personal information should be secret.