SaTC: EDU: AI for Cybersecurity Education via an LLM-enabled Security Knowledge Graph

Developing a skilled cybersecurity workforce is critical for national security in today’s digital age. Traditional education systems struggle to keep pace with emerging threats and diverse learning requirements. Cybersecurity education, involving complex tools and varied threat scenarios, requires a tailored, progressive learning approach to effectively cater to different skill levels. This project develops Artificial Intelligence (AI) tools for cybersecurity education using large language models (LLMs) augmented with a Security Knowledge Graph (AISecKG) to improve cybersecurity education. The project aims to (1) establish interactive teaching methods and design flexible and tailored learning strategies to suit the diverse needs of undergraduate, graduate, and professional students; and (2) enhance cybersecurity education in STEM by offering self-paced learning, personalized support, and extensive cybersecurity resources, with the assistance of generative AI, making it more accessible to a broad audience. This project introduces a novel, interdisciplinary approach to cybersecurity education. First, LLMs and cybersecurity knowledge graphs will be utilized to create interactive tools. These tools, such as chatbots, are designed for contextual learning and simulating cyber-attacks. Cybersecurity and AI experts will collaborate to design, validate, and tailor the cybersecurity content to cater to students at various learning stages. Leveraging LLMs and security knowledge graphs, the content will be regularly updated to reflect the latest cybersecurity trends and advancements. The interactive educational tools will engage the students with adaptive learning experiences, thereby improving accessibility and effectiveness of cybersecurity education. The AI and education experts will collaborate and use an AI-embedded metric system to assess students' cognitive engagement and measure the outcomes of their learning. This project will be structured as follows: (a) Develop a problem-based learning (PBL) curriculum focused on desired learning outcomes; (b) Develop evidence-based teaching modules within the Interactive-Constructive-Active-Passive (ICAP) learning framework for PBL cybersecurity education to emphasize student cognitive engagement in learning tasks, enhance student self-efficacy for navigating uncertain problems, and promote student learning outcomes; (c) Integrate learning and assessment modules with predictive analytics to identify the students at risk and provide appropriate and timely support for early intervention. Students' data security, privacy, and transparency will be ensured by designing ethical and explainable frameworks and responsible use of AI technologies in cybersecurity education. This project is supported by the Secure and Trustworthy Cyberspace (SaTC) program, which funds proposals that address cybersecurity and privacy, and in this case specifically cybersecurity education. The SaTC program aligns with the Federal Cybersecurity Research and Development Strategic Plan and the National Privacy Research Strategy to protect and preserve the growing social and economic benefits of cyber systems while ensuring security and privacy. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.