Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{182938,
        author = {Akhilesh Kumar},
        title = {Adversarial AI vs. Defensive AI: A Zero-Sum Game in Modern Cybersecurity},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {6},
        number = {10},
        pages = {430-432},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=182938},
        abstract = {The escalation of artificial intelligence (AI) in cybersecurity has led to an unprecedented arms race between adversarial and defensive AI. As malicious actors employ adversarial AI to bypass traditional and machine learning-based security mechanisms, defensive AI emerges to detect, respond to, and adapt against such intelligent threats. This paper investigates the competitive dynamics between adversarial AI and defensive AI within the framework of a zero-sum game, where gains by one agent imply direct losses to the other. The research explores real-world attack scenarios, including adversarial perturbations, AI-driven phishing, and data poisoning, countered by advanced defensive AI strategies such as anomaly detection, generative adversarial networks (GANs), and adversarial training. A comprehensive methodology involving simulation-based evaluation of threat models, countermeasures, and performance metrics is presented. Results indicate a constantly evolving equilibrium, where neither adversarial nor defensive AI achieves a permanent upper hand. The findings underscore the necessity for continuous learning architectures and AI governance frameworks. The paper concludes by advocating for a symbiotic human-AI collaboration and policy-driven AI ethics to mitigate the existential risks posed by adversarial threats.},
        keywords = {Adversarial AI, Defensive AI, Zero-Sum Game, Cybersecurity, Machine Learning Attacks, Adversarial Training, AI Ethics, Generative Adversarial Networks, Threat Detection, Cyber Warfare},
        month = {July},
        }