Collaborative Research: EAGER: SaTC-EDU: Learning Platform and Education Curriculum for Artificial Intelligence-Driven Socially-Relevant Cybersecurity

随着社交媒体的兴起，网络骚扰已被确定为一个关键的社会网络安全问题。人工智能（AI）具有通过自动检测来解决这一问题的巨大潜力。尽管人工智能可以成为打击网络骚扰的有用工具，但它很容易受到对抗性攻击。此外，人工智能驱动的网络骚扰检测模型可能会嵌入社会问题，例如公平和道德。为了推进人工智能网络安全教育，该项目将开发课程模块和动手实验室。这些模块将基于对人工智能驱动的网络骚扰检测、针对人工智能模型的相关攻击以及人工智能模型中的社会问题的前沿研究。该项目将有利于计算机科学（CS）和非CS（例如，社会科学）的学生，由于人工智能驱动的网络骚扰检测的高度跨学科性质。社交媒体时代网络骚扰的普遍性和严重性使该项目成为激励和教育学生了解人工智能和社交网络安全的相互需求和利益的理想选择。该项目将吸引具有不同背景的学生（特别是来自代表性不足的群体）进入人工智能网络安全领域，并提高对网络安全和人工智能的普遍认识。该项目的目标是将新兴社会网络安全的最新研究成果转化为教育形式。该项目将开发实践实验室，涵盖人工智能驱动的社会网络安全的不同方面，并展示人工智能和网络安全之间的相互作用。实践实验室将被集成到一个基于云的开放式学习平台中，该平台包含1）项目团队自主开发的经典AI驱动的网络骚扰检测算法; 2）针对这些AI算法和防御的对抗性攻击; 3）AI模型中的社会问题和偏见缓解。拟议的学习平台将通过自己的实验为学生提供对社会网络安全问题和人工智能技术的深入了解。项目团队将为计算机和非计算机学生开发课程材料，还将为高中生开发课程材料和组织夏令营，以提高他们对相关领域的网络安全意识和兴趣。该项目设计的基于云的开放实验室和学习平台将很容易被其他大学的教师访问。该项目得到了安全和值得信赖的网络空间（SaTC）计划的特别倡议的支持，以促进网络安全，人工智能和教育领域之间新的，以前未探索的合作。SATC计划与联邦网络安全研究和发展战略计划和国家隐私研究战略保持一致，以保护和维护网络系统日益增长的社会和经济效益，同时确保安全和隐私。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

With the rise of social media, cyberharassment has been identified as a critical social-cybersecurity problem. Artificial intelligence (AI) has immense potential to solve this problem by automating detection. Nevertheless, while AI can be a useful tool in the fight against cyberharassment, it is vulnerable to adversarial attacks. In addition, AI-driven cyberharrassment detection models may have embedded social problems, such as fairness and ethics. To advance AI-cybersecurity education, this project will develop curricular modules and hands-on labs. These modules will be based on cutting-edge research on AI-driven cyberharassment detection, related attacks against the AI models, and social issues in AI models for cyberharassment detection. This project will benefit both computer science (CS) and non-CS (e.g., social science) students due to the highly interdisciplinary nature of AI-driven cyberharassment detection . The pervasiveness and severity of cyberharassment in the era of social media makes this project ideal for motivating and educating students about the mutual needs and benefits of AI and social-cybersecurity. This project will attract students with diverse backgrounds (specifically from underrepresented groups) into the AI-cybersecurity field and increase general awareness of cybersecurity and AI. The goal of this project is to transform recent research outcomes in emerging social-cybersecurity into an educational format. This project will develop hands-on labs that cover different dimensions of AI-driven social-cybersecurity and demonstrate the interplay between AI and cybersecurity. The hands-on labs will be integrated into a cloud-based open learning platform, which contains 1) the project team’s homegrown and classic AI-driven cyberharassment detection algorithms; 2) adversarial attacks against these AI algorithms and defenses; and 3) social issues and bias mitigation in AI models. The proposed learning platform will provide students with an in-depth understanding of social-cybersecurity problems and AI techniques through their own experimentation. The project team will develop course materials for both CS and non-CS students and will also develop curriculum materials and organize summer camps for high school students to increase their cybersecurity awareness and interest in the related fields. The cloud-based open labs and learning platform designed in this project will be easily accessed by faculty from other universities. This project is supported by a special initiative of the Secure and Trustworthy Cyberspace (SaTC) program to foster new, previously unexplored, collaborations between the fields of cybersecurity, artificial intelligence, and 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.