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{187145,
        author = {Mrs. Sharon Spoorthy D S and Mrs. Sangeetha R and Shashidharan V and Priyanka V},
        title = {Artificial Intelligence Approaches for Next-Generation Cybersecurity Threat Detection: A Comprehensive Survey},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {3331-3347},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187145},
        abstract = {The escalating sophistication and frequency of cyber-attacks have necessitated the development of advanced detection and prevention mechanisms. Traditional security measures prove inadequate against modern threats such as zero-day exploits, advanced persistent threats (APTs), and polymorphic malware. This comprehensive survey examines the application of Artificial Intelligence (AI) techniques—specifically Machine Learning (ML), Deep Learning (DL), and metaheuristic algorithms—in detecting diverse cyber threats across multiple platforms including PCs, mobile devices, IoT systems, and cloud environments. Through systematic analysis of over sixty recent studies (2020-2024), we evaluate the effectiveness of AI-driven detection methods against malware, network intrusions, phishing attacks, ransomware, botnets, and insider threats. Our findings reveal that DL models achieve detection accuracies exceeding 99% on benchmark datasets, while metaheuristic algorithms significantly optimize feature selection and model performance. We propose a unified framework for assessing AI-based cybersecurity solutions and identify critical research gaps including cross-platform detection, adversarial robustness, and real-time deployment challenges. The study demonstrates that hybrid approaches combining multiple AI techniques offer superior performance compared to single-method solutions, with accuracies reaching 99.99% on datasets like CIC-IDS2018 and NSL-KDD. Our analysis emphasizes the imperative for continuous model evolution and adaptive learning systems to counter increasingly sophisticated attack vectors in modern cybersecurity landscapes.},
        keywords = {Artificial Intelligence, Machine Learning, Deep Learning, Cybersecurity, Intrusion Detection, Malware Detection, Metaheuristic Algorithms, Threat Intelligence.},
        month = {November},
        }