TIME
2024年12月24日（周二）14:00 – 15:00
VENUE
信息管理与工程学院308会议室
SPEAKER
Guangke Chen is a Research Assistant Researcher at Pengcheng National Laboratory. He earned his Ph.D. in 2024 from ShanghaiTech University as a recipient of the President’s Award and obtained his Bachelor's degree in 2019 from South China University of Technology. His research focuses on Trustworthy Artificial Intelligence, with 7 first-author papers published, including Big-4 security conferences such as IEEE S&P'21, USENIX Security'23, NDSS'24, and NDSS'25, as well as 2 papers in the flagship security journal IEEE TDSC, and 7 co-authored papers. His work has revealed security vulnerabilities in several commercial AI services and products, earning acknowledgments or rewards from vendors. His first-author paper published in IEEE S&P'21 has been cited over 240 times. He has been invited to serve as a program committee member or reviewer for 14 prestigious journals and conferences such as ICICS, INTERSPEECH, IEEE TDSC, IEEE TIFS, ACM TOPS, and Springer Cybersecurity.
陈光科，鹏城国家实验室助理研究员，于2024年在上海科技大学以校长奖学金获得者身份取得博士学位，于2019年在华南理工大学取得学士学位。他的研究方向是可信人工智能，目前以第一作者身份发表论文7篇，包括安全四大顶会IEEE S&P’21, USENIX Security’23, NDSS’24和NDSS’25, 安全旗舰期刊IEEE TDSC 2篇，以及其他合著论文7篇。他的研究揭示了多个商用人工智能服务或产品存在的安全漏洞，获得厂商致谢或奖励。他发表在IEEE S&P’21的第一作者论文被引240余次。他受邀担任ICICS, INTERSPEECH, IEEE TDSC, IEEE TIFS, ACM TOPS和Springer Cybersecurity在内的14个知名会议或期刊的程序委员会委员或审稿人。
TITLE
Security and Privacy of AI-based systems and AI-Generated Content Governance
ABSTRACT
Artificial Intelligence (AI) is revolutionizing industries and daily life but faces key challenges. AI-based systems’ outputs can be unreliable, breach privacy, or deviate from social norms. The misuse of generative AI confuses media authenticity and offenses humans through harmful synthetic content. These erode trust in AI and hinder its potential. In this talk, I will present my works addressing these challenges with two goals: (1) ensuring the security, privacy, and ethical standards of AI-based systems, and (2) mitigating risks from generative AI misuse, with a focus on audio/audio intelligent systems and multimodal AI.
编审：唐志皓 江波
欢迎 关注！
