SaTC: EDU: Inculcate a culture of preparedness against AI security threats to pervasive robotic systems

SaTC：EDU：灌输一种针对普遍机器人系统的人工智能安全威胁做好准备的文化

• 批准号: 2246920
• 负责人: Sudip Mittal
• 金额: $ 40万
• 依托单位: Mississippi State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-06-15 至 2026-05-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2246920&HistoricalAwards=false
• 关键词: SaTC; EDU; Inculcate; culture; preparedness

Robotics, automation, and related Artificial Intelligence (AI) systems have become pervasive, resulting in concerns related to security, safety, accuracy, and trust. With growing dependency on physical robots that work closely with humans, security is becoming increasingly important because cyber-attacks could lead to privacy invasion, critical operations sabotages, and bodily harm. The current shortfall of professionals who can defend such systems demands development and integration of new curricula. This project is creating a new curriculum with the scaffolding necessary to engage students at the junior/senior university level who are majoring in robotics, AI, and security related fields. More than 500 students are being directly served by this educational initiative at Mississippi State University. The project also indirectly impacts students across the nation at institutions in which faculty choose to integrate the modules developed for this project. On a broader scale, this project intends to increase public scientific literacy and engagement, especially in rapidly advancing fields such as cybersecurity, robotics, and AI. This is a significant societal benefit as it will improve the preparedness and well-being of individuals in a technologically advanced society and contribute to development of a diverse, globally competitive STEM workforce.The project is developing seven self-contained and adaptive modules on “AI security threats to pervasive robotic systems”. Topics will include: 1) introduction, examples of attacks, and motivation; 2) robotic AI attack surfaces and penetration testing; 3) attack patterns and security strategies for input sensors; 4) training attacks and associated security strategies; 5) inference attacks and associated security strategies; 6) actuator attacks and associated security strategies; and 7) ethics of AI, robotics, and cybersecurity. Course modules are evaluated to determine their impact on students. Workshops and tutorial sessions at conferences are used to expand the project’s impact and provide students and enthusiasts with hands-on experience. A two-day training workshop for external faculty further enables dissemination of the modules.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.

机器人、自动化和相关的人工智能(AI)系统已经变得无处不在，导致了与安全、安全、准确性和信任有关的担忧。随着对与人类密切合作的物理机器人的日益依赖，安全变得越来越重要，因为网络攻击可能导致隐私侵犯、关键操作破坏和身体伤害。目前缺乏能够捍卫这种制度的专业人员，因此需要开发和整合新的课程。该项目正在创建一个新的课程，为吸引主修机器人、人工智能和安全相关领域的初级/高级大学水平的学生提供必要的支架。密西西比州立大学的这项教育倡议直接服务于500多名学生。该项目还间接影响了全国各地的学生，在这些机构中，教职员工选择整合为该项目开发的模块。在更广泛的范围内，该项目旨在提高公众的科学素养和参与度，特别是在网络安全、机器人和人工智能等快速发展的领域。这是一项重大的社会效益，因为它将改善技术进步社会中个人的准备和福祉，并有助于发展一支多样化、具有全球竞争力的STEM劳动力队伍。该项目正在开发七个独立和自适应的模块，主题是“无处不在的机器人系统的人工智能安全威胁”。主题将包括：1)介绍，攻击的例子和动机；2)机器人AI攻击面和渗透测试；3)输入传感器的攻击模式和安全策略；4)培训攻击和相关的安全策略；5)推理攻击和相关的安全策略；6)执行器攻击和相关的安全策略；以及7)人工智能、机器人和网络安全的伦理。对课程模块进行评估，以确定它们对学生的影响。研讨会和会议上的辅导课程被用来扩大该项目的影响，并为学生和爱好者提供实践经验。为期两天的外部教师培训研讨会进一步促进了模块的传播。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。