Highlighting the Water Footprint of ChatGPT and its Impact on Usage Intention

Abstract

Although still rarely associated with it, the issue of sustainability is closely tied to generative AI, as its development and usage consume significant amounts of resources, particularly water. This study aims to explore how awareness of this issue affects individuals’ intended usage behavior of ChatGPT. The theory of planned behavior and Schwartz’s theory of basic human values are applied to explain usage intentions. Results from our experimental study reveal that information about ChatGPT’s high water consumption negatively impacts participants’ intention to use ChatGPT. Those who received the information reported a lower intention to use ChatGPT in the future compared to the control group, with this effect mediated by their attitude toward use. Additionally, the influence of information on attitude toward use was partially moderated by participants’ human values, particularly biospheric and hedonic values. Our findings contribute to a deeper understanding of user behavior in relation to generative AI. Furthermore, the results provide valuable insights both for policymakers who need to raise awareness for this issue, and for organizations that should integrate environmental aspects into their implementation strategies for generative AI in the workplace.

Ähnliche Arbeiten