Harmful Design in the Metaverse and How to Mitigate it: A Case Study of User-Generated Virtual Worlds on Roblox
Year: 2023
Type: conference-paper
Abstract: Metaverse platforms such as Roblox have become increasingly popular and profitable through a business model that relies on their end users to create and interact with user-generated virtual worlds (UGVWs). However, UGVWs are difficult to moderate, because game design is inherently more complex than static content such as text and images; and Roblox, a game platform targeted primarily at child players, is notorious for harmful user-generated game such as Nazi roleplay games and gambling-like mechanisms. To develop a better understanding of how harmful design is embedded in UGVWs, we conducted an empirical study to understand Roblox users’ experiences with harmful design. We identified several primary ways in which user-generated game designs can be harmful, ranging from directly injecting inappropriate content into the virtual environment of UGVWs to embedding problematic incentive mechanisms into the UGVWs. We further discuss opportunities and challenges for mitigating harmful designs.
Authors: Yubo Kou, Xinning Gui
Institution: Pennsylvania State University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Funder: National Science Foundation
OpenAlex ID: W4383681546
