Loading...
Thumbnail Image
Item

Visualizing AI Energy Use: Can Consumption Reminders Promote Social Good Mindsets?

Citations
Altmetric:
Soloist
Composer
Publisher
Date
2025-05-19
Additional date(s)
Abstract
Generative Artificial Intelligence (Gen AI) tools like ChatGPT are reshaping digital interactions across industries, but with significant, often invisible, environmental costs. From high energy consumption during model training and inference to large-scale water usage for data center cooling, Gen AI contributes to an expanding ecological footprint. Despite this, users remain largely unaware of their personal impact due to the seamless and abstracted nature of AI platforms. This study investigates whether integrating visual cues about energy usage into AI interfaces can raise awareness and promote more sustainable user behavior. Drawing on theories from environmental psychology, corporate social responsibility, and human-computer interaction, the research uses a mixed-methods experimental design involving 168 undergraduate university participants. Each participant interacted with a version of ChatGPT modified with different visual sustainability treatments, ranging from symbolic footnotes to interactive energy meters. While results did not show statistically significant changes in behavior across treatment groups, qualitative responses revealed increased reflection, emotional engagement, and openness to change, particularly among users exposed to subtler cues. These findings suggest that visibility alone is insufficient; effective intervention must be value-driven, emotionally resonant, and seamlessly integrated. The study contributes to the emerging field of ethical AI design by proposing that bridging the psychological distance between users and technological consumption is essential for promoting social good in the digital age.
Contents
Subject(s)
Research Projects
Organizational Units
Journal Issue
Genre
Description
Format
Department
Business Information Systems
Advisor
DOI