SoC Security Module for AI Product Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Artificial intelligence products are vulnerable to reproduction, replacement, and modification by attackers, leading to potential misuse and intellectual property rights violations, and existing security measures do not adequately protect these products, especially when used in electronic devices without separate security specifications.
Innovation Solution
A system on chip with a security module that receives and decodes an encoded artificial intelligence product using a double-encoded asset key and private key distributed via a key exchange scheme, ensuring secure AI operations by protecting the AI product from attackers and maintaining response speed without server load delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If AI products are deployed in electronic devices without separate security specifications, then ease of operation is improved, but security against attackers deteriorates
Solution Approach 1:
The patent implements nested security by placing an encoded AI product inside an encoded asset key, and the encoded asset key inside a double-encoded asset key structure. The security module decodes these nested layers sequentially using corresponding keys, creating a nested protection structure that maintains ease of operation while enhancing security against attackers.
Solution Approach 2:
The patent introduces an encoded asset key as an intermediary between the AI product and the AI accelerator. This intermediary layer prevents attackers from directly accessing or modifying the AI product, while the security module acts as a mediator to decode and verify the intermediary before allowing AI operations to proceed.
2Reliability
If double-encoded asset key and key exchange scheme are implemented, then security is improved, but device complexity increases
Solution Approach 1:
The patent merges the security verification and AI product decoding functions into a single integrated process within the security module. The same module that verifies the double-encoded asset key also decodes the AI product using the extracted asset key, eliminating the need for separate security hardware and reducing overall device complexity.
Solution Approach 2:
The security module performs self-service by automatically decoding the double-encoded asset key using the unique key, extracting the asset key, and then using that asset key to decode the AI product. This automated multi-step process eliminates manual security configuration and reduces operational complexity despite implementing advanced security measures.
3Reliability
If AI product is encoded and decoded using multiple keys, then intellectual property protection is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary encoding to the AI product and asset key before deployment. The AI product is pre-encoded with the asset key, and the asset key is pre-encoded with the unique key to create the double-encoded structure. This preliminary preparation enables rapid verification and decoding during operation, reducing processing time while maintaining strong IP protection.
Solution Approach 2:
The patent segments the encoding process into distinct layers: the AI product is encoded with the asset key, and the asset key is separately encoded with the unique key. This segmentation allows the security module to verify and decode each layer independently using corresponding keys, improving processing efficiency compared to a single complex encoding scheme.
Data Source
AI summary
Disclosed herein are a system on chip, a method of operating the system on chip, and an electronic device including the system on chip that execute artificial intelligence (AI) algorithms and/or machine learning algorithms in a 5G environment connected for Internet of Things in order to prevent an artificial intelligence product from being surreptitiously used, replaced, or modified by an attacker. The method for operating the system on chip may include a step of the system on chip receiving an encoded artificial intelligence product, a double-encoded asset key, and a private key distributed using a key exchange scheme, a step of receiving, from an electronic device including the system on chip, a unique key of the electronic device, a step of decoding the double-encoded asset key using the unique key of the electronic device in order to extract an encoded asset key, a step of decoding the encoded asset key using the private key in order to extract an asset key, a step of decoding the encoded artificial intelligence product using the asset key in order to extract an artificial intelligence product, and a step of performing artificial intelligence operation using the artificial intelligence product. It is possible to prevent an artificial intelligence product made by an individual, a company, or an organization from being reproduced and surreptitiously used, replaced, or arbitrarily modified by an attacker.