A Proactive AI Framework for Detecting and Mitigating Mutating Malware Generated via Generative AI Models: A Literature Review

Abstract

The cybersecurity landscape has changed due to the rapid development of generative artificial intelligence (AI), which empower attackers to generate highly adaptive malware that can bypass currently available detection methods. This study reviews current research (2023–2026) on initiative-taking AI-driven strategies done on purpose to identify and get rid of such real time threats. It looks at advancements in behavioural analytics, polymorphic malware, adversarial machine learning, and predictive threat intelligence. The study underlines the need for flexible, intelligence-driven defence mechanisms and draws attention to the diminishing efficacy of signature-based systems. Important discoveries show that to successfully fight AI-generated malware, future cybersecurity solutions must integrate behavioural monitoring, ensemble learning, and anticipatory threat modelling. In the paper's conclusion, research gaps are listed and strategies for developing robust, next-generation detection frameworks are suggested.