AI-Enhanced Protocol Fuzzing: Integrating Machine Learning with Defensics for Advanced Vulnerability Detection
Year: 2025
Type: article
Abstract: This article presents an innovative approach to protocol fuzzing by integrating artificial intelligence capabilities with the Defensics security testing platform. The article addresses critical challenges in modern protocol security testing, particularly focusing on the complexities of evolving communication protocols and their diverse implementation landscapes. The proposed AI-enhanced framework introduces advanced machine learning techniques for optimizing test case generation, improving vulnerability detection rates, and streamlining resource utilization across various protocol implementations. Through comprehensive experimental validation and real-world case studies, the research demonstrates significant improvements in testing efficiency, coverage metrics, and vulnerability detection capabilities compared to traditional fuzzing approaches. The framework incorporates dynamic learning mechanisms, adaptive testing strategies, and sophisticated resource allocation algorithms to enhance the overall effectiveness of security testing processes. This integration of AI capabilities with established fuzzing methodologies represents a significant advancement in automated security testing, offering improved protocol coverage and more efficient vulnerability detection across diverse deployment scenarios.
Author: Gurdeep Kaur Gill
Institution: Cisco Systems (United States)
Language: English
Subfield: Software
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4407562314