Machine learning algorithms are in the process of reshaping the ways in which Internet users consume content and interact with each other online. Meanwhile, the algorithm’s ability to generate and manipulate media content also revolutionizes the way how attackers run deceptive campaigns that aim to manipulate public options or perform social engineering attacks. In this talk, I will describe our recent efforts to understand the emerging threat vectors introduced by machine learning algorithms and study how they would affect users. First, I will talk about how machine learning can be used to subvert common tools designed for cybercrime forensics and tracing sources of misinformation, using perceptual hashing based image search as an example. Second, I will share our study results to understand how users perceive (and establish trust towards) algorithm-generated online personas in the social engineering context. Finally, I will discuss the open questions in combating online deception and the role that machine learning can play to strengthen our defenses.
Gang Wang is an Assistant Professor of Computer Science at University of Illinois at Urbana-Champaign. He obtained his Ph.D. from UC Santa Barbara in 2016, and a B.E. from Tsinghua University in 2010. Before joining University of Illinois, he worked as an assistant professor at Virginia Tech from 2016 to 2019. His research interests are Security and Privacy, Data Mining, and Internet Measurements. His work primarily takes a data-driven approach to address emerging security threats in massive communication systems (e.g., online social networks, email systems), crowdsourcing systems, mobile applications, and enterprise networks. He is a recipient of the NSF CAREER Award (2018), Amazon Research Award (2021), Google Faculty Research Award (2017), and Best Paper Awards from IMWUT 2019, ACM CCS 2018, and SIGMETRICS 2013. His projects have been covered by various media outlets such as MIT Technology Review, The New York Times, Boston Globe, and ACM TechNews.