1. How Institutional Defaults Shape Data Center Regulation in Australia and South Korea

• In Event: Infrastructure, Industrial Policy, and the Political Economy of AI

Abstract

As artificial intelligence drives unprecedented demand for data center infrastructure globally, governments face urgent decisions about how to govern the physical backbone of the digital economy. Yet regulatory responses vary dramatically across jurisdictions, and not simply because of different policy preferences or levels of digital development. This paper examines why South Korea and Australia, two comparable high-income Asia-Pacific democracies with significant digital economies, have developed fundamentally different approaches to data center governance, and what this divergence means for how each country will respond to the accelerating infrastructure demands of the AI era.

Drawing on the complete population of regulatory documents pertaining to data centers in both countries, comprising 34 Korean documents and 105 Australian documents, we use topic modeling and network analysis alongside qualitative discourse analysis to map the regulatory landscape of data center governance in each jurisdiction. Our findings reveal a striking and consequential divergence. South Korea has constructed data centers as a distinct national technology asset, embedding them within an explicit framework of state promotion, technological investment, and AI-oriented industrial policy. Australia, by contrast, has no equivalent regulatory identity for data centers. They are governed implicitly through pre-existing land use, planning, and infrastructure frameworks, without a coherent national strategy that positions them as strategic AI infrastructure. Crucially, because our corpus constitutes the complete population of relevant regulatory documents in both countries rather than a sample, the absence of data-center-specific regulatory language in Australia is itself a substantive finding about how that country has institutionally approached digital infrastructure governance.

We argue that this divergence reflects not merely different policy choices but deeply embedded patterns in each country's governance tradition that determine how new technologies get categorised and managed before deliberate policy decisions are even made. South Korea's tradition of constructing explicit regulatory frameworks around strategic technologies has produced a coherent, nationally coordinated imaginary of data centers as instruments of technological advancement. Australia's federated planning tradition has absorbed data centers into existing frameworks without generating an equivalent national vision.

These findings carry direct implications for AI governance. Countries that have not explicitly constructed data centers as strategic AI infrastructure face a structural governance gap as AI demand accelerates. Addressing that gap requires not just new policies but a deliberate national reimagining of what data centers are and what they are for. Without this, federated and consent-based planning systems risk being fundamentally misaligned with the speed and scale of AI infrastructure investment now underway across the Asia-Pacific region. We conclude with reflections on how policymakers can recognise and address these institutional patterns before they constrain AI infrastructure capacity and national competitiveness in the AI era.

Authors

• Seokkyun Woo, Korea Advanced Institute of Science and Technology
  Non-Presenting Co-Author

• Jeonghyun Seo, Korea Advanced Institute of Science and Technology
  Presenting Author