The Methodological Footprint of Generative AI

• In Event: Complicated and Complicating Questions About the Role of AI in Qualitative Inquiry

Abstract

As generative AI tools become increasingly accessible, qualitative researchers are experimenting with their integration into inquiry processes—to navigate large datasets, make sense of patterns, or support coding (e.g., Morgan, 2023; 2025). While these tools offer opportunities to reimagine how qualitative inquirers engage with data, their use is not without consequence (Authors, 2025). In this paper, we take up a critical and timely concern: the environmental costs of incorporating generative AI into qualitative research and the methodological footprints left behind.

Recently, scholars have urged us to consider the ethical and epistemological implications of AI in education and research (Davison, et al., 2024; Gillen, 2024). Less attention, however, has been paid to the material and ecological tolls. Drawing upon Author et al.’s, (2024) discussion of methodological footprints, this paper brings environmental questions to the fore. They (2024) argued “it is essential that qualitative scholars consider the ecological footprint and broader impact of their work and how these forces … could be anticipated, addressed, reduced, and questioned” (p. 485). Similarly, we argue that engaging AI in qualitative research requires not only methodological reflexivity, but ecological accountability and ecological intelligence (Goleman, 2009). If qualitative research is rooted in care, relationality, and attentiveness to context (Puig de la Bellacasa, 2017), then our methodological choices must account for the broader systems in which our tools are embedded—including their environmental impacts.

The proliferation of large language models requires enormous computing power, contributing to rising energy consumption and electronic waste (Luccioni et al., 2024). The demand for “data lakes” to train these models also requires physical and digital storage infrastructures whose environmental costs remain largely hidden (see e.g., Meerman et al, 2023). These impacts are not peripheral to the practice of qualitative research; rather, they are entangled with the very tools we now use to know and represent the social world.

Moreover, as institutions of higher education call for “innovation” and “efficiency”, the uptake of AI is often framed as necessary, inevitable, and even virtuous (Nartey, 2025). Yet as Fielding and Lee (2008) reminded us, “…new digital technologies come with strings attached in the form of research policies, institutional expectations, and shifting boundaries around customary norms … in qualitative research” (p. 491). Now, those strings include a deepening complicity in systems that exacerbate ecological degradation. Consequently, scholars must be concerned about the ecological cost and sustainability of their practices.

Thus, we expand the scope of methodological reflexivity toward ecologies and sustainability: to consider not only how we produce knowledge, but at what environmental cost. What does it mean to adopt generative AI tools in an era of climate crisis? What are the long-term implications of scaling up our research with tools that contribute to ecological harm? And how might we cultivate forms of inquiry that reduce, rather than deepen, our methodological footprints, in terms of environmental and epistemological justice? Rather than offering easy answers, we call for deeper, ongoing conversations about responsibility, sustainability, justice, and care in qualitative research in the age of AI.

Authors

• Jessica Nina Lester, Indiana University

• Jessica Van Cleave, Gardner-Webb University

• Mirka E. Koro, Arizona State University