AI Sovereignty: The Next Frontier of Global Power

As the global race for artificial intelligence (AI) accelerates, a new form of geopolitical and economic competition is emerging: AI sovereignty. AI sovereignty is broader than data residency alone. It encompasses control over the infrastructure, governance frameworks, compute power and data ecosystems that underpin AI innovation. While discussions around sovereignty have traditionally focused on data residency and governance, the conversation must now expand to include control over computing infrastructure, data ecosystems and the ability to innovate at scale. Nations that fail to act decisively risk becoming structurally dependent on a handful of dominant players,  most notably the United States, which is rapidly consolidating its leadership in AI  infrastructure.

A growing imbalance in global compute power

The scale of that dominance is hard to ignore. The United States currently operates more than 5,000 data centers, with an additional 3,000 more being planned; this is more than the rest of the world combined. This imbalance creates a profound asymmetry in access to the computational backbone required to power AI. Because AI is inherently compute intensive, countries without sufficient domestic infrastructure face a binary choice:  constrain their ambitions or outsource their AI workloads to foreign providers.

That dependency comes with real and immediate risks. Outsourcing compute limits innovation velocity, and organizations also face latency, cost pressures, and regulatory friction. More critically, reliance on foreign infrastructure introduces systemic exposure.  In an environment defined by shifting geopolitical dynamics, access to essential compute resources could become a lever of influence, restricting not just growth, but sovereignty itself.

For highly connected economies such as Singapore, the challenge is particularly nuanced. The country depends heavily on cross-border digital flows, meaning policymakers must strike a delicate balance between maintaining openness and preserving control over sensitive national and enterprise data.

The data sovereignty paradox

But compute is only half the equation. Data is the lifeblood of AI, and it introduces an equally complex set of challenges. Many countries have enacted stringent data sovereignty regulations that mandate keeping sensitive data within national borders.  Yet if compute resides elsewhere, how can those countries reconcile compliance with their AI ambitions? Moving data across borders increases exposure to privacy risks, governance challenges and foreign legal frameworks. Retaining data within national boundaries is not just about compliance; it is about ensuring that citizen data is not repurposed to train external models or subjected to jurisdictional overreach.

Singapore has taken a governance-led approach to this challenge. Through initiatives such as its Trusted Data Sharing Framework and National AI Strategy, the country is attempting to accelerate AI innovation while maintaining tighter oversight of how sensitive data is shared and governed.

The economic stakes of AI leadership

The macroeconomic implications are equally profound. AI is expected to expand from contributing just 0.3% of U.S. GDP today to more than 15% within the next decade. This represents one of the largest redistributions of economic value in modern history.  Countries that are constrained in how they operationalize AI will inevitably limit their share of this opportunity. The result is a widening gap in technological capability as well as economic competitiveness and national prosperity.

Realizing this gap, Singapore has taken a similarly aggressive approach. Under its  National AI Strategy 2.0, the government committed more than SGD1 billion to strengthen  AI capabilities, infrastructure and talent development as part of its ambition to become a leading global AI hub. However, this level of investment might not be enough.

Building modern AI infrastructure is capital-intensive, with hyperscale data centers in the U.S. costing upwards of USD5 billion each. Technology giants are investing at unprecedented levels. Microsoft alone has committed more than US$200 billion to expand its data center footprint. Even large national investments pale in comparison.  India, for example, has allocated approximately US$200 billion toward AI development,  underscoring the scale of the global investment gap.

From dependence to sovereignty: A path forward

However, there is a viable path forward. AI sovereignty is achievable, but it requires intentional investment and a rethinking of national strategy. Countries must prioritize the development of domestic data center infrastructure, leveraging regional advantages such as lower energy costs and labor efficiencies to build at scale. We are already seeing early signs of this shift.

Projects such as Singapore’s Jurong Island Mega Park infrastructure initiatives reflect a broader regional push to build AI-ready ecosystems that combine compute capacity, energy resilience and stricter data governance. Yet Singapore also illustrates the complexity of AI sovereignty. Data centers already account for roughly 7% of the country’s electricity consumption, forcing policymakers to balance AI expansion against sustainability, land constraints and long-term energy resilience.

However, infrastructure alone will not unlock the full value of AI. Nations must also address the fragmentation of their data ecosystems. Today, critical data remains siloed across organizations and systems, limiting its usability for AI. A logical approach to data integration, one that enables secure, governed access to distributed data with zero-copy,  offers a powerful alternative. By adopting such a strategy, organizations can accelerate  AI adoption while remaining compliant with sovereignty requirements.

As McKinsey has noted, “CDOs who want to scale AI throughout their organizations will  need to treat data as a core enterprise asset.” That principle also applies to AI proper,  and at a national level as well. The countries that succeed will be those that treat data not as a constraint, but as a strategic asset that is unified, accessible and governed.

Ultimately, AI sovereignty is not about isolation but about control, resilience and growth.  AI sovereignty is no longer just about protecting data. It is increasingly about determining where innovation happens, who captures economic value and which nations retain strategic control in the AI era. By investing in domestic infrastructure,  securing data assets and enabling seamless access to information, nations can reduce external dependency while fostering innovation from within. The countries that act now will protect their digital futures and define the next era of global economic leadership.

The views and opinions expressed in this article are those of the author and do not necessarily reflect those of CDOTrends. Image credit: iStockphoto/ALLVISIONN