AI -Driven Cryptographic Algorithm Identification: Exploring Methodologies and Practical Applications
Shivani Yadao, Nguyễn Thị Diệu Linh, Areesha Fatima, Bhuvan Puri · 2024
This paper examines the application of Artificial Intelligence (AI) and Machine Learning (ML) techniques for classifying and detecting cryptographic algorithms in network traffic, particularly as cyber threats and financial scams continue to rise globally. Cryptographic algorithms are fundamental to data security, ensuring confidentiality, integrity, and authentication. The study explores the evolving role of AI in cryptography, focusing on its potential to overcome limitations of traditional detection methods, particularly in addressing novel threats and encrypted data. It analyzes how these advanced approaches can efficiently classify and identify encryption algorithms, potentially revolutionizing cybersecurity practices. The paper discusses practical applications of AI/ML in cryptographic algorithm identification, examining their impact on malicious code detection, real-time threat analysis, and overall cryptographic security enhancement. Additionally, it addresses current challenges in the field and proposes future research directions. Using data extracted from CrossRef, this study provides a comprehensive analysis of current research trends and methodologies in AI-driven cryptographic identification. By leveraging AI's capacity to analyze large datasets and adapt to unknown algorithms, this approach aims to significantly improve the reliability and efficiency of cybersecurity systems. The study contributes to the evolving landscape of cryptographic security.