*Raymond Ternenge Igbudu1, Shaibu John Ojonugwa1, Gilbert Imuetin Osaze Aimufua1, Magaji Yusuf1
1Center for Cyberspace Studies, Nasarawa State University Keffi
*Corresponding author’s Email: igbuduraymond@gmail.com, doi.org/10.55639/607.020100141
ABSTRACT
The growing complexity, frequency, and sophistication of cyber threats have increased the need for adaptive and data-driven cybersecurity capabilities. This paper presents a structured literature review of artificial intelligence (AI) applications, challenges, and future directions in cybersecurity. The review covered literature published from 2019 to 2025 and searched four sources: IEEE Xplore, ScienceDirect, SpringerLink, and Google Scholar. After reconciling the submitted reference list, 58 sources were retained: 52 journal articles, one conference paper, three technical reports, and two preprints. The review synthesizes evidence on machine learning (ML), deep learning (DL), natural language processing (NLP), intrusion detection, malware analysis, vulnerability assessment, threat intelligence, Internet of Things (IoT) security, explainable AI (XAI), adversarial machine learning, autonomous security operations centres, and quantum-related cybersecurity concerns. The findings indicate that AI can improve anomaly detection, threat classification, predictive analysis, and security automation, but its effectiveness depends on data quality, model robustness, interpretability, privacy protection, and appropriate human oversight. The review also identifies dual-use risks, algorithmic bias, accountability concerns, and adversarial manipulation as persistent challenges. Real-world case evidence is considered cautiously because some reported performance improvements originate from vendor or industry sources rather than independent evaluations. The review concludes that future AI-enabled cybersecurity research should emphasize robust and explainable models, privacy-preserving learning, adversarial resilience, human-AI collaboration, trustworthy governance, and representative benchmark datasets.
KEYWORDS
Artificial
Intelligence,
Cybersecurity,
Machine
Learning,
Deep Learning,
IoT Security.