G2C Icon G2C Icon

Welcome to G2C, a gallery of AI infrastructure.

Ground to Cloud (G2C) brings together scholarly research and investigative journalism on the vast array of infrastructure that supports contemporary artificial intelligence, with a focus on specific sites and forms of labor. Each of the six layers on the left collects examples of the material configurations that underlie the supposedly immaterial cloud, often involving very real extraction and exploitation. From cobalt mining in the Congo to logjams in the Panama Canal to image labelling in Venezuela, these exhibits serve as a corrective to the sleek, futuristic, and conspicuously depeopled images of neatly arranged server racks that dominate the official imagery of AI.

Another aim of G2C is to demonstrate the astonishing scale and breadth of AI systems. It is no accident that the writers cited throughout make liberal use of the term "planetary"—researchers have examined the "planetary extractive machine" (Labban 2014),

Mazen Labban, “Deterritorializing Extraction: Bioaccumulation and the Planetary Mine,” Annals of the Association of American Geographers 104, no. 3 (2014): 560–76, https://www.jstor.org/stable/24537757.
"planetary-scale computation" (Bratton 2015),
Benjamin H. Bratton, The Stack: On Software and Sovereignty, Software Studies (Cambridge, Mass. London: MIT press, 2015).
"planetary labor markets" (Graham 2019),
Mark Graham and Mohammad Amir Anwar, “The Global Gig Economy: Towards a Planetary Labour Market?,” First Monday, April 1, 2019, https://doi.org/10.5210/fm.v24i4.9913.
and the "planetary costs of AI" (Crawford 2021).
Kate Crawford, The Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence (Yale University Press, 2021), https://doi.org/10.2307/j.ctv1ghv45t.
Artificial intelligence requires a truly earth-spanning infrastructure, one that collides and overlaps in different ways with older forms of power and sovereignty. My hope is that the exhibits here can offer a point of departure for exploring the kinds of site-specific interactions that result.

Text highlighted in teal

Bibliographic information will appear in a popup. Click to visit the bibliography on Zotero.
indicates a citation; hover or focus to preview and click to see complete bibliographic information on Zotero. You can return to this page at any time by selecting the Ground to Cloud Favicon icon at the top left.

G2C is built and maintained by Owen Leonard and hosted on Github Pages. A full bibliography for the site is available on Zotero. The code is open-source and can be viewed on Github. Suggested citation for this site:

Owen Leonard, "Ground to Cloud: A Gallery of AI Infrastructure" (2025), https://g2c.owenleonard.dev/