Adaptive DNA Cryptography With Intelligent Machine Learning for Cloud Data Defense

Abstract

ABSTRACT With the exponential growth of sensitive data generated and stored in cloud environments, traditional cryptographic techniques are increasingly strained in addressing emerging security threats. While DNA cryptography offers promising security potential due to its complexity and biological uniqueness, its standalone application lacks adaptive intelligence for real‐time threat mitigation. This research addresses the gap in integrating biologically inspired encryption with artificial intelligence by proposing a hybrid framework that combines DNA‐based cryptographic encoding with machine learning for robust cloud data protection. The proposed methodology introduces an adaptive model where DNA cryptographic operations—such as DNA encoding, XOR operations, and complementary rules—are optimized using supervised machine learning algorithms. These algorithms dynamically enhance encryption‐decryption processes based on anomaly detection and performance metrics. Experimental results on benchmark cloud datasets demonstrate significant improvements in encryption strength, key management, and resistance to common attacks, without compromising processing efficiency. The implications suggest a novel direction for future‐proofing cloud security infrastructures through the convergence of bio‐computing and intelligent systems.

Keywords

• cloud security
• cryptosystem
• DNA cryptography
• genetics cryptography