Paper Type
Complete
Paper Number
1660
Description
This study addresses the pressing need to mitigate algorithmic biases in human Generative Artificial Intelligence (GenAI) collaboration, given the inherent uncertainties and biases in AI-generated insights. Leveraging the Protection Motivation Theory (PMT), this work proposes the concept of individual bias mitigation motivation and behavior. Then, a research model was proposed to shed light on the drivers of such motivation and behavior from users. An empirical investigation using structural equation modeling (N=198) reveals that users' perceived severity, self-efficacy, and response efficacy significantly influence their motivation to mitigate biases. Subsequently, users' motivation positively impacts their bias mitigation behavior, thereby enhancing their task performance. This study elucidates the mechanism underlying bias mitigation in human-GenAI collaboration, offering valuable insights for both academia and industry. By introducing the concept of bias mitigation motivation and behavior and employing PMT, this study enriches the discourse on human-AI symbiosis and provides actionable guidance for optimizing collaboration processes.
Recommended Citation
Huynh, Minh-Tay, "Users’ Bias Mitigation Behavior in Human-Generative AI Collaboration: A Protection Motivation Theory Perspective" (2024). PACIS 2024 Proceedings. 19.
https://aisel.aisnet.org/pacis2024/track13_hcinteract/track13_hcinteract/19
Users’ Bias Mitigation Behavior in Human-Generative AI Collaboration: A Protection Motivation Theory Perspective
This study addresses the pressing need to mitigate algorithmic biases in human Generative Artificial Intelligence (GenAI) collaboration, given the inherent uncertainties and biases in AI-generated insights. Leveraging the Protection Motivation Theory (PMT), this work proposes the concept of individual bias mitigation motivation and behavior. Then, a research model was proposed to shed light on the drivers of such motivation and behavior from users. An empirical investigation using structural equation modeling (N=198) reveals that users' perceived severity, self-efficacy, and response efficacy significantly influence their motivation to mitigate biases. Subsequently, users' motivation positively impacts their bias mitigation behavior, thereby enhancing their task performance. This study elucidates the mechanism underlying bias mitigation in human-GenAI collaboration, offering valuable insights for both academia and industry. By introducing the concept of bias mitigation motivation and behavior and employing PMT, this study enriches the discourse on human-AI symbiosis and provides actionable guidance for optimizing collaboration processes.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
Interaction