Clemson team develops safeguards for AI-generated youth content

An interdisciplinary team led by Clemson University is launching research to make generative AI safer for young people, focusing on content that can be sexually explicit, abusive, toxic or otherwise not safe for work. Researchers said such material can affect mental health and has been linked to anxiety, cyberbullying, isolation and eating disorders.

The project will examine how generative AI interprets prompts and creates unsafe content, then develop safeguards that can detect hidden intent, remove unsafe elements from images and review generated images and videos. The team also plans to survey and interview young adults and parents to assess whether the tools reduce harmful exposure and influence how people use and respond to AI.

Linke Guo, an associate professor in Clemson’s Holcombe Department of Electrical and Computer Engineering, is the principal investigator. The Clemson-led group includes researchers in computer science, electrical and computer engineering, sociology and criminology, with collaboration from the University of Texas at Arlington. The four-year effort is supported by $1.2 million from the National Science Foundation through its Secure & Trustworthy Cyberspace program.