Investigating Moderation Challenges to Combating Hate and Harassment: The Case of Mod-Admin Power Dynamics and Feature Misuse on Reddit
Madiha Tabassum, Alana Mackey, Ashley Schuett, Ada Lerner
33rd USENIX Security Symposium · Day 1 · USENIX Security '24
Online platforms continue to grapple with pervasive issues of hate, harassment, and abuse, which manifest in diverse forms from targeted hate speech to doxing and non-consensual sharing of intimate media. This talk, presented at USENIX Security '24, delves into the intricate challenges faced by volunteer community moderators on Reddit, who serve as the frontline defense against such harms. The research, a collaboration between Wesley College, George Washington University, and Northeastern University, critically examines two core issues: the complex and often fraught relationship between volunteer moderators and Reddit's paid platform administrators, and the adversarial misuse of benign platform features to facilitate malicious activities.
AI review
This research forcefully argues for treating platform feature misuse as a technical security problem, not just a social one. By applying threat modeling and a security mindset to social design, the talk exposes how benign features can be weaponized, offering a crucial paradigm shift for platform builders and policy makers. The demo is a stark, effective illustration of systemic failure.