Online Archive of University of Virginia Scholarship
AI vs AI: Detecting AI-Generated and Scripted Social Media Content; AI-Generated Misinformation in the Gaza-Israeli War110 views
Author
Zhang, Kevin, School of Engineering and Applied Science, University of Virginia
Advisors
Norton, Peter, EN-Engineering and Society, University of Virginia
Morrison, Briana, EN-Comp Science Dept, University of Virginia
Abstract
How can we maximize the benefits of generative artificial intelligence (AI) but minimize the harm?
How can machine learning combat the proliferation of bots and generative AI on social media? As up to 50 percent of social media accounts are run by bots, an efficient means of detecting such accounts is necessary. To interpret text, a machine learning model is proposed that would apply natural language processing techniques, such as word vectorization and sentiment lexicons. By identifying and searching for text patterns associated with AI and bot accounts, the model would classify account content as man-made, botted, or AI-generated.
How is generative AI tampering with the online landscape? In the Gaza War, fighting occurs both on the battlefield and online. Through bot networks, AI-generated images, and AI comments and messages, both sides have filled cyberspace with manipulative and deceptive messages. Designed to provoke emotion, users spread such content, unintentionally amplifying propaganda.
Degree
BS (Bachelor of Science)
Keywords
Generative AI; Misinformation; Machine Learning; Natural Language Processing
Notes
School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Briana Morrison
STS Advisor: Peter Norton
Technical Team Members: Kevin Zhang
Zhang, Kevin. AI vs AI: Detecting AI-Generated and Scripted Social Media Content; AI-Generated Misinformation in the Gaza-Israeli War. University of Virginia, School of Engineering and Applied Science, BS (Bachelor of Science), 2025-12-12, https://doi.org/10.18130/akgj-y025.
