AI Encryption Code Detection for Unauthorized Access Prevention

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing AI systems face challenges in quickly and accurately identifying new malicious programs and preventing unauthorized data encryption, as they rely on signature-based methods that are inefficient against emerging threats and cannot effectively block encryption beyond approved rights.

Innovation Solution

An AI-powered electronic device and method that uses a machine learning-based encryption code identification model to detect and block encryption algorithms within executable code, executing applications in a secure environment and preventing unauthorized encryption attempts by hooking memory loads and updating vulnerability lists.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If signature-based methods are used to identify malicious programs, then detection accuracy for known threats is improved, but response time and effectiveness against new malicious programs deteriorate

Engineering Contradiction:
Improvedetection accuracyVSAvoidresponse time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of executable code by extracting features and inputting them into a machine learning model before the program executes. This advance detection mechanism identifies malicious programs including new threats without waiting for signature updates, resolving the contradiction between detection accuracy and response time.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If traditional malicious program identification methods are used, then removal or blocking of identified programs is achieved, but prevention of encryption beyond approved rights deteriorates

Engineering Contradiction:
Improvemalicious program removalVSAvoidencryption rights control
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system monitors program execution and provides feedback by continuously analyzing code behavior against approved encryption rights. When a program attempts encryption beyond its authorized scope, the system detects this through the machine learning model and blocks the operation, enabling adaptive control of encryption rights while maintaining reliable malicious program removal.

Inventive Principle:
Principle #23Feedback

3Device complexity

If AI technology is not used, then system simplicity is maintained, but ability to detect encryption vulnerability exploits deteriorates

Engineering Contradiction:
Improvesystem simplicityVSAvoidencryption exploit detection
Core Design Contradiction:
Device complexityVSDifficulty of detecting and measuring

Solution Approach 1:

The system replaces traditional mechanical signature-matching methods with an AI-based machine learning model that automatically analyzes code features and detects encryption algorithms. This substitution significantly improves the ability to detect encryption vulnerability exploits and new malicious programs, accepting increased system complexity as necessary for enhanced security detection capabilities.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3665604B1Security enhancement method and electronic device therefor
Publication Date: 2023.02.15 SAMSUNG ELECTRONICS CO LTD
  • EP3665604B1 patent drawingFigure 1
  • EP3665604B1 patent drawingFigure 2A
  • EP3665604B1 patent drawingFigure 2B

AI summary

An electronic device and method that are robust against attacks on encryption-related vulnerabilities as detection of an encryption algorithm based on if artificial intelligence technology is enabled are provided. A security enhancement method includes a hooking loading of an executable code into a memory, inputting the executable code into an encryption code identification model that is based on an artificial neural network, determining, by the encryption code identification model, whether the loading of the executable code into the memory is allowed, and when the loading of the executable code is not allowed, blocking the loading of the executable code into the memory.