Jaehong Kim

Jaehong Kim

Moral Emotion · Attention Economy · Machine Behavior

I am a Ph.D. candidate at the Korea Advanced Institute of Science and Technology (KAIST), advised by sociologist Wonjae Lee. I completed two research internships at the Max Planck Institute for Security and Privacy (MPI-SP), supervised by computational social scientist Meeyoung Cha, with whom I continue to collaborate. My research is supported by the Hyundai Motor Chung Mong-Koo Foundation. Beyond academia, my research on foreign-linked influence operations has received national media coverage, including an SBS News report with over 1.2 million views, and has contributed to policy discussions among lawmakers.

Google Scholar LinkedIn CV Email

News

Research

My research examines how moral emotions shape human and machine behavior.

Human Behavior. I build multimodal frameworks that detect moral emotions in English and Korean text and images. My findings show that moral outrage (anger at perceived wrongdoing) most strongly predicts committed user participation and is disproportionately used by suspected foreign-linked trolls.

Machine Behavior. I study how large language models internally represent moral outrage. Steering these representations causally shifts punishment judgments, leading models to recommend harsher sentences for identical offenses.

Publications

Ongoing Work

Education