AI-Driven Offensive and Defensive Security for Web and IoT Systems

The rapid convergence of web applications, cloud-native services, and Internet of Things (IoT) ecosystems has fundamentally reshaped modern communication infrastructures. Web APIs, microservices architectures, edge computing, and IoT protocols such as MQTT and CoAP now form highly interconnected, distributed systems that operate across heterogeneous networks. While this transformation has enabled scalability and flexibility, it has also expanded the attack surface dramatically. Simultaneously, Artificial Intelligence (AI) and Machine Learning (ML) techniques are transforming cybersecurity practice. AI is now used to automate vulnerability discovery, conduct intelligent fuzzing, generate exploit payloads, analyze firmware, and perform behavioral anomaly detection. Large language models (LLMs) have further accelerated automated code auditing, exploit synthesis, and security misconfiguration detection. However, AI is also being weaponized—enabling adversarial attacks, automated reconnaissance, and adaptive exploitation strategies.This Research Topic seeks to explore the dual role of AI in both offensive and defensive security for distributed web and IoT systems, with particular emphasis on communication-layer vulnerabilities, authentication mechanisms, and protocol-level security.Scope and Objectives: This Research Topic aims to advance scientific understanding and methodological rigor in AI-enabled security assessment and defense. Contributions should focus on communication-centric security challenges, including protocol exploitation, authentication weaknesses, distributed trust models, and real-time detection in networked environments.Encourage submissions that:i. Propose novel AI-driven methodologies for vulnerability discovery in web and IoT systems.ii. Investigate automated exploit generation and adversarial testing frameworks.iii. Develop adaptive defense mechanisms leveraging ML for intrusion detection, anomaly detection, or self-healing architectures.iv. Provide formal models or empirical evaluations of AI-driven penetration testing systems.v. Examine communication overhead, scalability, and robustness of AI-based security frameworks. Topic editors: Akashdeep Bhardwaj, Kuo-Hui Yeh, Antonino Galletta Source: https://www.frontiersin.org/research-topics/78914/ai-driven-offensive-and-defensive-security-for-web-and-iot-systems