Abstract

Artificial intelligence and machine learning are increasingly embedded in modern cybersecurity systems because of their capacity to automate threat detection, analysis, and anomaly identification. This shift toward intelligent defense has, however, introduced a new class of vulnerabilities, collectively termed adversarial machine learning (AML), in which attackers craft malicious inputs, poison training data, or mount inference-based attacks to subvert model behavior. This study reviews the emerging adversarial threats targeting cybersecurity applications and critically appraises the defense strategies proposed to strengthen model robustness and resilience. A structured literature review of 22 peer-reviewed studies published between 2022 and 2025 was conducted to identify dominant attack patterns, the most vulnerable application domains, and the limitations of current defenses. To ground the review in measurable evidence, a reproducible case study is additionally reported in which a neural intrusion-detection model is subjected to a Fast Gradient Sign Method (FGSM) evasion attack on the NSL-KDD dataset and then hardened through adversarial training. The undefended detector's accuracy collapses from 80.8% to near 0% as the perturbation budget grows, whereas the adversarial trained detector retains roughly 70-75% accuracy under the same attack, at the cost of a small reduction in clean accuracy. The findings confirm that adversarial attacks are becoming increasingly sophisticated, particularly against intrusion detection systems, autonomous systems, and deep-learning-based security tools, and that, although adversarial training, defensive distillation, and explainable AI are promising, open questions remain regarding their scalability, adaptability, and real-time applicability. The study underscores the need for multi-layered, adaptive security strategies to enhance the trustworthiness of AI-powered cybersecurity solutions.

References

• Ejeofobiri, C. K., Fadare, A. A., Fagbo, O. O., Ejiofor, V. O., and Fabusoro, A. T. 2024. The role of artificial intelligence in enhancing cybersecurity: A comprehensive review of threat detection, response, and prevention techniques. International Journal of Science and Research Archive, 13(2), 310-316.
• Ali, S., Wang, J., and Leung, V. C. M. 2025. AI-driven fusion with cybersecurity: Exploring current trends, advanced techniques, future directions, and policy implications for evolving paradigms - A comprehensive review. Information Fusion, 118.
• Malik, J., Muthalagu, R., and Pawar, P. M. 2024. A systematic review of adversarial machine learning attacks, defensive controls, and technologies. IEEE Access, 12, 99382-99421.
• Zhou, S., Liu, C., Ye, D., Zhu, T., Zhou, W., and Yu, P. S. 2023. Adversarial attacks and defenses in deep learning: From a perspective of cybersecurity. ACM Computing Surveys, 55(8).
• Pelekis, S., Koutroubas, T., Blika, A., Berdelis, A., Karakolis, E., Ntanos, C., and Askounis, D. 2025. Adversarial machine learning: A review of methods, tools, and critical industry sectors. Artificial Intelligence Review, 58(8).
• Alotaibi, A., and Rassam, M. A. 2023. Adversarial machine learning attacks against intrusion detection systems: A survey on strategies and defense. Future Internet, 15(2).
• Girdhar, M., Hong, J., and Moore, J. 2023. Cybersecurity of autonomous vehicles: A systematic literature review of adversarial attacks and defense models. IEEE Open Journal of Vehicular Technology, 4, 417-437.
• Adesina, D., Hsieh, C. C., Sagduyu, Y. E., and Qian, L. 2023. Adversarial machine learning in wireless communications using RF data: A review. IEEE Communications Surveys and Tutorials, 25(1), 77-100.
• Vaccari, I., Carlevaro, A., Narteni, S., Cambiaso, E., and Mongelli, M. 2022. explainable and reliable against adversarial machine learning in data analytics. IEEE Access, 10, 83949-83970.
• Barik, K., Misra, S., and Lopez-Baldominos, I. 2025. Black-box adversarial attack defense approach: An empirical analysis from cybersecurity perspective. Results in Engineering, 26.
• Apruzzese, G., Andreolini, M., Ferretti, L., Marchetti, M., and Colajanni, M. 2022. Modeling realistic adversarial attacks against network intrusion detection systems. Digital Threats: Research and Practice, 3(3).
• Merkle, F., Samsinger, M., Schoettle, P., and Pevny, T. 2024. On the economics of adversarial machine learning. IEEE Transactions on Information Forensics and Security, 19, 4670-4685.
• Kiranbabu, M. N. V., Jeraldine Viji, A., Chandanan, A. K., Birchha, V., Pandey, T. K., and Sar, S. K. 2025. The challenge of adversarial attacks on AI-driven cybersecurity systems. Journal of Cybersecurity and Information Management, 15(1), 288-297.
• McCarthy, A., Ghadafi, E., Andriotis, P., and Legg, P. 2022. Functionality-preserving adversarial machine learning for robust classification in cybersecurity and intrusion detection domains: A survey. Journal of Cybersecurity and Privacy, 2(1), 154-190.
• Ododo, F. R., and Sadiq, R. R. 2025. Adversarial attacks in cybersecurity: A machine learning perspective. Journal of Science Innovation and Technology Research.
• Ennaji, S., De Gaspari, F., Hitaj, D., Kbidi, A., and Mancini, L. V. 2025. Adversarial challenges in network intrusion detection systems: Research insights and future prospects. IEEE Access, 13, 148613-148645.
• Barik, K., and Misra, S. 2024. Adversarial attack defense analysis: An empirical approach in cybersecurity perspective. Software Impacts, 21.
• Khan, M., and Ghafoor, L. 2024. Adversarial machine learning in the context of network security: Challenges and solutions. Journal of Computational Intelligence and Robotics, 4(1), 51-63.
• Ke, H., Xu, J., Wang, Y., Chen, H., and Shen, Z. 2025. Adversarial machine learning in cybersecurity: Attacks and defenses. International Journal of Management Science Research, 8(2), 26-33.
• Paya, A., Arroni, S., Garcia-Diaz, V., and Gomez, A. 2024. Apollon: A robust defense system against adversarial machine learning attacks in intrusion detection systems. Computers and Security, 136.
• Jiang, T., Liu, Y., and Cui, X. 2025. Textual adversarial attacks in cybersecurity named entity recognition. Computers and Security, 150.
• Alsmadi, I., Aljaafari, N., Nazzal, M., Alhamed, S., Sawalmeh, A. H., Vizcarra, C. P., and Al-Humam, A. 2022. Adversarial machine learning in text processing: A literature survey. IEEE Access, 10, 17043-17077.
• Goodfellow, I. J., Shlens, J., and Szegedy, C. 2015. Explaining and harnessing adversarial examples. In Proceedings of the International Conference on Learning Representations (ICLR). arXiv:1412.6572.
• Madry, A., Makelov, A., Schmidt, L., Tsipras, D., and Vladu, A. 2018. Towards deep learning models resistant to adversarial attacks. In Proceedings of the International Conference on Learning Representations (ICLR). arXiv:1706.06083.
• Tavallaee, M., Bagheri, E., Lu, W., and Ghorbani, A. A. 2009. A detailed analysis of the KDD CUP 99 data set. In Proceedings of the IEEE Symposium on Computational Intelligence for Security and Defense Applications (CISDA), 1-6.