Location
Hilton Hawaiian Village, Honolulu, Hawaii
Event Website
https://hicss.hawaii.edu/
Start Date
3-1-2024 12:00 AM
End Date
6-1-2024 12:00 AM
Description
Artificial intelligence (AI) is increasingly incorporated into innovative personal health apps to improve the decision-making of its users. To facilitate the understanding and to increase usage of such AI-based personal health apps, firms are progressively turning to explainable artificial intelligence (XAI) designs. However, we argue that explanations of the AI-based recommendations have not only positive but also negative consequences. Based on a socio-technical lens, we develop a model that relates XAI to technostress - both eustress and distress - and its downstream consequences. To test our model, we conducted an online experiment, in which participants interact with XAI or black-box AI. Our results show that (1) XAI causes both eu- and distress, and (2) simultaneously exerts differential influence on objective performance, satisfaction, and intention to use. Our findings contribute to information systems research and practice by uncovering the dual effect of XAI on decision processes in the health context.
Recommended Citation
Grüning, Maximilian; Wolf, Tobias; and Trenz, Manuel, "A Stressful Explanation: The Dual Effect of Explainable Artificial Intelligence in Personal Health Management" (2024). Hawaii International Conference on System Sciences 2024 (HICSS-57). 2.
https://aisel.aisnet.org/hicss-57/hc/wellness_management/2
A Stressful Explanation: The Dual Effect of Explainable Artificial Intelligence in Personal Health Management
Hilton Hawaiian Village, Honolulu, Hawaii
Artificial intelligence (AI) is increasingly incorporated into innovative personal health apps to improve the decision-making of its users. To facilitate the understanding and to increase usage of such AI-based personal health apps, firms are progressively turning to explainable artificial intelligence (XAI) designs. However, we argue that explanations of the AI-based recommendations have not only positive but also negative consequences. Based on a socio-technical lens, we develop a model that relates XAI to technostress - both eustress and distress - and its downstream consequences. To test our model, we conducted an online experiment, in which participants interact with XAI or black-box AI. Our results show that (1) XAI causes both eu- and distress, and (2) simultaneously exerts differential influence on objective performance, satisfaction, and intention to use. Our findings contribute to information systems research and practice by uncovering the dual effect of XAI on decision processes in the health context.
https://aisel.aisnet.org/hicss-57/hc/wellness_management/2