Search

Bridging the Practice Gap: An AI-Integrated Cybersecurity Pedagogy for Developing Operational Readiness

Author(s)

Li Song*, Li Jianfeng, Zeng Haowen

Affiliation(s)

School of Information Engineering, Nanning College of Technology, Guilin , China *Corresponding Author

Abstract

As cyber threats have become increasingly sophisticated in 2025, the gap between theoretical knowledge and practical application in cybersecurity education has widened into a critical chasm. Traditional teaching methods, often reliant on static textbooks and predictable laboratory exercises, frequently produce graduates with a condition best described as "strong theory but weak hands-on skills." These students, while well-versed in concepts, struggle to apply their knowledge in dynamic, real-world scenarios, leaving them unprepared for the modern security operations center. This paper proposes a novel pedagogical framework integrated with Artificial Intelligence, termed the "AI-Cyber-Praxis" framework, designed specifically to address this skills gap. By leveraging specialized, fine-tuned Large Language Models as personalized, 24/7 pedagogical tutors and utilizing advanced AI-driven Automated Penetration Testing tools within highly dynamic virtual cyber ranges, the proposed model creates an adaptive, high-intensity, and deeply engaging learning environment. This framework shifts the traditional passive teaching mode, compelling students into active, real-time confrontation with intelligent, evolving threats. The research demonstrates that the integration of AI-enhanced feedback loops, Socratic-style guidance, and personalized challenge generation significantly improves students' critical thinking, problem-solving capabilities, and technical proficiency in core cybersecurity competencies.

Keywords

AI-Enhanced Cybersecurity Education; Large Language Models; Adaptive Cyber Ranges; Hands-On Skill Development; Automated Penetration Testing

References

[1] Lingzhi Wang, Xinyi Shi, Ziyu Li, et al, “Automated Penetration Testing with LLM Agents and Classical Planning”, 2025, arXiv:2512.11143. [2] Gholami, Yazi . "Large Language Models (LLMs) for Cybersecurity: A Systematic Review." World Journal of Advanced Engineering Technology and Sciences 13.1(2024):057-069. [3] Lei Wu, Xiaofeng Zhong, Jingju Liu, and Xiang Wang. PTGroup: An Automated Penetration Testing Framework Using LLMs and Multiple Prompt Chains. In Advanced Intelligent Computing Technology and Applications: 20th International Conference, ICIC 2024, Tianjin, China, August 5–8, 2024, Proceedings, Part IX. Springer-Verlag, Berlin, Heidelberg, 220–232. [4] Luis Javier García Villalba. "Analysis of Autonomous Penetration Testing Through Reinforcement Learning and Recommender Systems." Sensors 25(2025). [5] Deng, Gelei, et al. "{PentestGPT}: Evaluating and harnessing large language models for automated penetration testing." 33rd USENIX Security Symposium (USENIX Security 24). 2024. [6] Hannay, Jo Erskine, Audun Stolpe, and Muhammad Mudassar Yamin. "Toward ai-based scenario management for cyber range training." International Conference on Human-Computer Interaction. Cham: Springer International Publishing, 2021. [7] Umaroh, Liya, Mukaromah Mukaromah, and Muhammad Naufal. "Enhancing speaking skill through ai-powered technology." Seminar Nasional Teknologi Dan Multidisiplin Ilmu (SEMNASTEKMU). Vol. 3. No. 1. 2023. [8] Andreolini, Mauro, et al. "A framework for the evaluation of trainee performance in cyber range exercises." Mobile Networks and Applications 25.1 (2020): 236-247. [9] Chopra S, Ahmad H, Goel D, et al. Chatnvd: Advancing cybersecurity vulnerability assessment with large language models[J]. arXiv preprint arXiv:2412.04756, 2024