This paper proposes a novel Genetic Algorithm (GA)-based code obfuscation technique using Abstract Syntax Trees (ASTs) to enhance software security. The method aims to protect proprietary logic from reverse engineering by generating diverse obfuscated code variants. It applies variable renaming, dead code insertion, and control flow changes within a GA framework, optimized for interpreted languages like Python. A multi-objective fitness function evaluates both cyclomatic complexity and execution time to balance obfuscation strength and performance. Experimental results show that the technique significantly increases code complexity while preserving functionality. The approach demonstrates strong potential for securing software against unauthorized analysis, offering an effective defense through intelligent, language-aware code transformation.
Copyrights © 2025