Large Language Models for Cybersecurity Education: A Survey of Current Practices and Future Directions

Nan Sun, Yuantian Miao, Xiaoxing Mo, Jun Zhang

Abstract: The increasing complexity of cyber threats demands innovative approaches to cybersecurity education that overcome the limitations of traditional teaching methods. This survey paper presents the first comprehensive review of Large Language Models (LLMs) applications in cybersecurity education, examining how these advanced AI systems can address current pedagogical challenges. While existing surveys have explored LLMs in general education or specific cybersecurity applications, our work uniquely focuses on the intersection of LLMs and cybersecurity education. We analyze empirical studies demonstrating LLMs’ effectiveness in delivering interactive, personalized learning experiences that adapt to individual student needs. The survey examines current implementations that leverage LLMs’ capabilities to create dynamic training materials, provide real-time feedback, and simulate real-world scenarios. We particularly emphasize how LLMs can offer scalable, cost-effective solutions that make cybersecurity education more accessible while maintaining currency with evolving threats. The paper concludes by identifying promising future directions for LLM integration in cybersecurity education, providing valuable insights for educators, researchers, and curriculum developers working to enhance cybersecurity training frameworks.

External IDs:dblp:conf/pakdd/SunMMZ25