Vol. 1 No. 2 (2021): African Journal of Artificial Intelligence and Sustainable Development
Articles

Towards Autonomous Security: Leveraging Artificial Intelligence for Dynamic Policy Formulation and Continuous Compliance Enforcement in Zero Trust Security Architectures

Mahammad Shaik
Technical Lead - Software Application Development, Charles Schwab, Austin, Texas, USA
Leeladhar Gudala
Associate Architect, Virtusa, New York, USA
Cover

Published 23-07-2021

Keywords

  • Zero Trust Security,
  • Artificial Intelligence,
  • Machine Learning,
  • Natural Language Processing,
  • Security Policy Management

How to Cite

[1]
Mahammad Shaik and Leeladhar Gudala, “Towards Autonomous Security: Leveraging Artificial Intelligence for Dynamic Policy Formulation and Continuous Compliance Enforcement in Zero Trust Security Architectures”, African J. of Artificial Int. and Sust. Dev., vol. 1, no. 2, pp. 1–31, Jul. 2021, Accessed: Dec. 25, 2024. [Online]. Available: https://africansciencegroup.com/index.php/AJAISD/article/view/28

Abstract

The cyber threat landscape is characterized by continuous innovation on the part of adversaries, rendering traditional perimeter-based security models increasingly ineffective. Zero Trust Security (ZTS) architectures emerge as a response to this challenge, mandating rigorous verification of all access requests and enforcing the principle of least privilege. However, the dynamic nature of ZTS, with its emphasis on continuous evaluation and context-aware access control, presents significant challenges in formulating and enforcing security policies. These challenges encompass the need for scalability to accommodate evolving user bases and system configurations, as well as the ability to adapt to novel threats and attack vectors. This research investigates the potential of Artificial Intelligence (AI) to augment ZTS by automating policy formulation and implementing continuous compliance monitoring.

We explore how AI techniques, such as machine learning (ML) algorithms, can be harnessed to analyze vast datasets encompassing user behavior patterns, system activity logs, and threat intelligence feeds. Through supervised learning approaches, AI models can be trained to identify normal access patterns and resource utilization behaviors. Deviations from established baselines can then trigger alerts, enabling the dynamic generation of context-aware security policies. This allows for granular access control that adapts to user roles, device characteristics, and the specific context of access requests. Furthermore, natural language processing (NLP) techniques can be employed to extract insights from security policies expressed in human-readable formats. This facilitates the translation of these policies into machine-interpretable rules, enabling automated policy enforcement and configuration across diverse IT infrastructure components.

The paper delves further into the application of AI for real-time anomaly detection within ZTS environments. By employing unsupervised learning algorithms, AI systems can analyze network traffic, system logs, and user activity for patterns that deviate from established baselines. This enables the proactive identification of potential security breaches, such as lateral movement attempts or unauthorized access attempts. The efficacy of AI-driven ZTS policies is evaluated through the development of a theoretical framework that emphasizes the ability to achieve dynamic adaptation to the ever-changing threat landscape. The research concludes by outlining potential limitations associated with the adoption of AI in ZTS, such as the explainability of AI-generated policies and the challenges of mitigating bias within training data. Additionally, the paper highlights promising avenues for future research, including the exploration of explainable AI (XAI) techniques to enhance transparency in policy decision-making, and the development of federated learning approaches to address privacy concerns in the context of threat intelligence sharing.

Downloads

References

  1. A. Ghare et al., "Zero trust security - Model and implementation," 2010 International Conference on Advanced Information Networking and Applications (AINA), Singapore, 2010, pp. 1018-1023, doi: 10.1109/AINA.2010.187.
  2. Y. Mao et al., "A Survey of Cloud Computing Security: Systems and Applications," Journal of Network and Computer Applications, vol. 94, pp. 1-16, 2017, doi: 10.1016/j.jnca.2017.04.001
  3. N. Papernot et al., "Distillation as a Defense to Adversarial Perturbations in Deep Learning," in Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, Vienna, Austria, 2016, pp. 1462-1477, doi: 10.1145/2937864.2941458
  4. I. Goodfellow et al., "Explaining and Attributing Decisions of Machine Learning Systems," arXiv preprint arXiv:1606.08853, 2016.
  5. P. McDaniel et al., "Federated Learning for Privacy-Preserving Fraud Detection," in 2019 IEEE International Conference on Big Data (Big Data), Los Angeles, CA, USA, 2019, pp. 5239-5246, doi: 10.1109/BigData.2019.8908102.
  6. H. Cheng et al., "Improving Federated Learning Performance: Communication Efficiency and Privacy," in Proceedings of the 2020 ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '20, New York, NY, USA, 2020, pp. 4300-4310, doi: 10.1145/3394480.3400173.
  7. V. Vapnik, "The Nature of Statistical Learning Theory," Springer Science & Business Media, 1995.
  8. L. Wasserman, "All of Statistics: A Concise Course in Statistical Inference," Springer Science & Business Media, 2004.
  9. I. Goodfellow et al., "Deep Learning," MIT Press, 2016.
  10. Y. LeCun et al., "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86(11), pp. 2278-2324, 1998, doi: 10.1109/5.726791.
  11. D. E. Rumelhart et al., "Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Volume 1: Foundations," MIT Press, 1986.
  12. J. Schmidhuber, "Neural Networks for Long-Term Memory," Istituto Nazionale per l'Alta Matematica Francesco Severi, 2005.
  13. S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, vol. 9(8), pp. 1735-1780, 1997, doi: 10.1162/neco.1000093.
  14. V. Nair and G. Hinton, "Recurrent Neural Networks for Recognition and Paragraph Generation," arXiv preprint arXiv:1409.0478, 2014.
  15. D. P. Kingma and M. Welling, "Auto-Encoding Variational Bayes," arXiv preprint arXiv:1312.6114, 2013.
  16. J. An et al., "Unsupervised Anomaly Detection with LSTMs for Industrial Big Data," IEEE Transactions on Industrial Informatics, vol. 13.