Generative artificial intelligence (GAI) is becoming widespread, to the point that we do not always recognize when we are using it, for instance in Google searches or when we use enterprise software suites that “come with GAI embedded,” such as Microsoft 365. Organizations of various kinds also use GAI for internal management, customer-facing processes, and more.
GAI can surely increase efficiency and automation in organizations (e.g., for routine tasks), the public sector (e.g., urban planning), and for private citizens (e.g., GAI agents). However, among the various concerns associated with GAI, one relates to the environment. The rapid development of LLMs required to power GAI has exacerbated these concerns to the next level. Training and running these systems require substantial amounts of electricity and natural resources, notably water for cooling, to operate and continue to improve (Davison & Marabelli, 2025). Furthermore, the demand for ML chips has a major environmental impact due to their high level of obsolescence, the resulting e-waste, and the extraction of minerals that are currently scarce resources. In addition, these impacts are uneven too. Global warming is already changing the conditions in which we live, especially in regions prone to extreme heat or rapid temperature changes. Moreover, people (and countries) at the margins have fewer defenses against rising temperatures: these people (and countries, generally belonging to the Global South) typically benefit the least from technology (Islam & Winkler, 2017). Research indicates that GAI is increasingly contributing to climate change, even if the magnitude of its impact remains unclear (Luccioni et al., 2025).
All this said, GAI can also benefit the environment if we refer to advances in weather and wildfire forecasting, smart cities, smart agriculture, etc. (Marabelli & Davison, 2025). Yet, the costs are gradually becoming more visible. For example, the cost of maintaining these technologies is outweighing the human resources, leading to major cutbacks and redundancies of professional staff. This Special Issue invites authors to adopt an eco-socio-technical perspective in the age of GAI (Watson, 2024).
With this call for papers, we aim to publish studies that examine the dual effect of GAI on the environment and explore how this technology can be implemented and used responsibly by individuals, organizations, communities, and governments. We strongly encourage submissions with primary data, even though we will consider submissions leveraging secondary data and, in exceptional cases, conceptual and review submissions. All methods typically used in IS journals are allowed (qualitative, quantitative, mixed, experiments, design science, perceptual studies, etc.). Practitioner papers (see ISJ paper genres) are welcomed.