Special session 10:
Hyperscale Geographies

Advancements in artificial intelligence (AI) are increasingly driven by scaling laws—the principle that greater accumulation of data, computational power, and infrastructure generates exponential improvements in performance. Hyperscalers—dominant cloud service providers such as Amazon, Microsoft, Google, Tencent, Huawei, and Alibaba—supply the computational infrastructure that powers AI and other data-intensive industries. Their dominance extends beyond technical infrastructure by reshaping the political economy of global capitalism, reinforcing dependencies among firms, states, and markets and prompting pressing questions regarding sovereignty, governance, and territorial control, especially in the context of growing geopolitical tension and the unfolding polycrisis.

​

Drawing on Benjamin Bratton’s concept of the stack, this panel explores how hyperscaler infrastructure reorganizes political and economic power across scales. The stack is not only a technical model of computational architecture but also a political and geoeconomic structure that shapes the circulation of goods, information, and capital. Geographers’ long-standing engagement with scale—as both an analytic lens and an empirical problem—offers valuable tools for theorizing the stack. From relational and multi-scalar methodologies to theories of scalar fixes, geography can address how the political and economic power embedded in hyperscaler infrastructure reorganizes the territorial and political arrangements between states, markets, and firms.

​

We invite papers that examine how hyperscaler stacks reshape sovereignty and global economic asymmetries, the uneven geographies of hyperscaler control and dependency, and the technical and geopolitical contestations surrounding cloud infrastructure. We particularly encourage contributions that explore how different sectors and regions shape the political economy of hyperscalers.