Call for Papers: Locating Power and Resistance Within Global AI Production Networks

Language:

Call for Papers

American Association of Geographers (AAG) 2026 Annual Meeting

Session Organizers

Mark Graham, Oxford Internet Institute, University of Oxford (mark.graham@oii.ox.ac.uk)
Uma Rani, International Labour Organization (amara@ilo.org)

Session Description

Artificial intelligence is not simply code running in a distant cloud; it rests on a sprawling production network that reaches from cobalt mines in the Democratic Republic of Congo to wafer fabrication plants in Taiwan, and from data-labeling offices in Nairobi to hyperscale data centers in the American West. The development and benefits of frontier technologies like AI are highly concentrated, with the United States and China accounting for the vast majority of funding for AI start-ups. This concentration of power in the hands of a few countries and corporations deepens existing global inequalities

Along these global AI value chains distinct harms accumulate. The environmental impact is staggering, with the information and communications technology (ICT) sector’s greenhouse gas emissions growing rapidly with the uptake of AI, and it includes a massive demand for electricity and water for its data infrastructures that can deepen climate injustices. Furthermore, the production of AI hardware relies on critical minerals, often sourced under precarious conditions, linking the digital transition to the same extractive patterns as the green transition.

The human cost is equally severe, with precarious and often invisible labor practices which are endemic to this industry. While AI-driven automation is projected to impact 40% of jobs worldwide, the consequences will be unevenly distributed. Developing countries are less exposed to job losses from automation but are also far less likely to reap the productivity benefits, potentially widening the gap between developed and developing nations. This dynamic creates a risk where developing countries provide raw materials and low-cost labor for AI production while the value and profits are captured elsewhere. These harms fall unevenly on workers, communities, and ecosystems while value and profit concentrate elsewhere.

This session invites papers that both map these harms and locate leverage points for change within AI production networks. There is a pressing need for international cooperation to create a global AI framework that prioritises people over technology and ensures that the benefits of AI are equitably distributed and we welcome empirical, methodological, and conceptual work that shows where interventions are possible, whether through policy, labor organizing, technical standards, alternative ownership structures, or public accountability campaigns. Key, but not exhaustive, questions include:

Where are the critical nodes? Which firms, jurisdictions, or infrastructural chokepoints could reduce resource extraction, carbon intensity, or exploitative subcontracting?
How can transparency travel upstream? What mapping or audit methods reveal hidden flows of minerals, capital, data, and labor, and how might visibility translate into enforceable accountability, particularly when few developing countries have national AI strategies?
What regulatory or procurement levers exist? How can state action reshape incentives along the chain, for example export controls on GPUs or government cloud contracts?
Where do labor and community struggles gain traction? We seek examples such as annotation-worker organizing, climate-tech alliances, Indigenous land claims, or right-to-repair movements that have shifted practices and challenge the trend of automation favoring capital over labour.
How can design choices reduce harm? Possible areas include hardware-efficient models, privacy-preserving data regimes, and open-source governance that embeds fairness and ecological limits, moving towards a more circular digital economy.

By foregrounding both the geographies of harm and the loci of agency, the session aims to build a collective agenda for transforming AI production toward more just and sustainable futures. Contributions that blend critical mapping, network analysis, ethnography, policy analysis, and activist scholarship are especially encouraged, as are perspectives attentive to the intersecting axes of power shaping AI’s global production landscape.

Submission Guidelines

Please submit an abstract of up to 250 words that clearly addresses the session theme, outlines methods and findings (where applicable), and situates the work within broader debates on the geographies of technology and production. Interdisciplinary approaches spanning economic geography, development studies, STS, labour studies, environmental geography, and critical data studies are especially encouraged. We welcome submissions from scholars at all career stages.

Deadline for abstracts to session organisers: Oct 14
Format: Please send your abstract (as a Word attachment) and a short bio (max 100 words) to mark.graham@oii.ox.ac.uk and amara@ilo.org with the subject line “AAG 2026 AI Production Networks Session”.