Year: 2024
Type: conference-paper
Abstract: This paper explores the potential of computer games as a new form of explainable interface to AI. Most existing eXplainable AI (XAI) provide explanations in static or limited interactive forms. This paper analyzes simulation games as an exploratory case study based on an established taxonomy in human-centered XAI. Our analysis indicates that the existing mechanics and game interfaces of simulation games, to various extents, can support most XAI question types, although certain XAI ques- tion types are difficult to convey. We offer initial reflections on the design space of leveraging insights from games to rethink the explainable interfaces for XAI.
Authors: Jennifer Villareale, Thomas Fox, Jichen Zhu
Institutions: Drexel University, IT University of Copenhagen
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4408260210