Authors:
Hadi Hussein Madhi,Ali Dahir Alramadan,DOI NO:
https://doi.org/10.26782/jmcms.2026.02.00005Keywords:
Cloud Security,Artificial Intelligence,Cryptography,Hybrid Framework,Intrusion Detection,AES-ECC Encryption,Adaptive Threat Detection,Cybersecurity,Information Asymmetry,Deep Learning.,Abstract
The exponential growth of cloud computing has brought both operational efficiency and complex cybersecurity challenges. Traditional intrusion detection systems (IDS) struggle to adapt to dynamic attack patterns and ensure data confidentiality. This research proposes a hybrid Artificial Intelligence–Cryptographic Framework that integrates deep learning and lightweight encryption to achieve adaptive threat detection while maintaining secure communication within cloud environments. Using the CICIDS 2023 and UNSW-NB15 datasets, the model combines a CNN–LSTM network for behavioral anomaly recognition with AES–ECC encryption for data integrity. Experimental results show a detection accuracy of 98.2 %, an F1-score of 97.9 %, and a 50 % reduction in false positives compared with traditional AI models, while maintaining an average encryption latency of 45 ms. Statistical validation using the Wilcoxon signed-rank test confirmed the significance of these improvements (p < 0.05). The study contributes theoretically by bridging information asymmetry, signaling, and fair-value principles into cybersecurity and practically by providing a scalable, efficient, and trust-aware solution for adaptive cloud protection.Refference:
I. Ali, S. (2025). Security and privacy in multi-cloud and hybrid systems. Journal of Cloud Security, 12(3), 45–60.
II. Alazab, M., Alazab, M., & Zhang, J. (2023). AI-driven intrusion detection in cloud environments. Computers & Security, 127, 103056.
III. Alazab, M., Alhyari, S., Awajan, A., & Abdalla, A. (2023). Machine learning-based intrusion detection systems in cloud computing. Computers & Security, 125, 103028.
IV. Alshamrani, M., Bahashwan, A., & Alotaibi, B. (2020). Machine learning techniques for cybersecurity threat detection: A comprehensive review. IEEE Access, 8, 221990–222010.
V. Ahmad, N., & Javed, H. (2023). Hybrid AI–blockchain frameworks for reliable cloud security. Journal of Information Security Research, 12(4), 210–225.
VI. Current Study (2025) refers to the authors’ ongoing research and therefore is not externally published.
VII. Deegan, C. (2022). Fair Value Theory and Its Role in Enhancing Corporate Reporting Transparency. Accounting Perspectives, 18(1), 33–49.
VIII. Hassan, M., Noor, M., & Rahim, R. (2024). Integrating AES and LSTM models for adaptive cloud threat mitigation. Computers & Security, 132, 103355.
IX. Kaur, P., & Singh, S. (2021). Deep learning-based intrusion detection framework using CNN–LSTM model. Future Generation Computer Systems, 115, 225–238.
X. Kim, Y., Park, H., & Seo, J. (2024). Cognitive CNN–LSTM-based intrusion detection for virtualized cloud environments. Expert Systems with Applications, 242, 121816.
XI. Rahman, M., Chowdhury, S., & Alam, K. (2022). Blockchain and AI-enabled hybrid systems for secure cloud infrastructures. IEEE Transactions on Cloud Computing, 10(6), 3624–3637.
XII. Rahman, M., Chowdhury, F., & Zhang, T. (2022). Benchmarking hybrid AI models for adaptive threat detection. Cybersecurity (SpringerOpen), 5(3), 18–32.
XIII. Smith, J., & Jones, A. (2022). Modern cloud architecture and threats. International Journal of Cloud Computing, 9(1), 1–20.
XIV. Zhao, L., Chen, Y., & Li, H. (2022). Federated learning architectures for privacy-preserving cloud intrusion detection. Information Sciences, 603, 112–128.
XV. Zhou, W., Li, P., & Wang, X. (2024). Information asymmetry and trust in AI-driven security frameworks. Journal of Information Technology, 39(2), 211–228.
View Download