Publication Details

Download

Beat Rossmy, Sarah Theres Völkel, Elias Naphausen, Patricia Kimm, Alexander Wiethoff, Andreas Muxel
Punishable AI: Examining Users' Attitude Towards Robot Punishment
In Proceedings of the 2020 ACM Designing Interactive Systems Conference (DIS '20). Association for Computing Machinery, New York, NY, USA, 179-191. DOI:https://doi.org/10.1145/3357236.3395542 (bib)

To give robots, which are black box systems for most users, feedback we have to implement interaction paradigms that users understand and accept, for example reward and punishment. In this paper we present the first HRI experience prototype which implements gradual destructive interaction, namely breaking a robot's leg as a punishment technique. We conducted an exploratory experiment (N=20) to investigate participants' behavior during the execution of three punishment techniques. Using a structured analysis of videos and interviews, we provide in-depth insights into participants' attitude towards these techniques. Participants preferred more abstract techniques and felt uncomfortable during human-like punishment interaction. Based on our findings, we raise questions how human-like technologies should be designed. A video documentation of the project can be found here: https://vimeo.com/348646727