July 9, 2026
By Atuf Guliyev
Key Takeaways
AI Infrastructure Has Become a Geopolitical Asset — Control over data centers, energy, chips, and networks is now central to great-power competition, determining economic competitiveness, military strength, and global influence.
Europe Lags Behind the US and China — While strong in regulation (AI Act), Europe is heavily dependent on US hyperscalers, faces energy and investment constraints, and struggles to build sovereign AI infrastructure.
Europe Needs a Realistic Path to Strategic Autonomy — It cannot fully replicate the US private-sector model or China’s state-driven approach. Instead, it should focus on sustainable computing, secure sovereign systems, and a “framework nations” strategy where countries like France, Germany, and the Nordics lead in specific areas.
How Europe can navigate the new geopolitics of compute infrastructure
The field of artificial intelligence has gone beyond just technology. It is now a key part of great-power competition, with physical infrastructure—like data centers, energy supplies, semiconductor chips, and ultra-fast networks—gaining geopolitical importance. In 2026, developing and controlling this infrastructure will reshape the global landscape, affecting alliances, industrial policies, energy strategies, and even the idea of strategic independence. In this context, artificial intelligence offers both opportunities and risks for the European Union, which could lose its strategic independence and become a follower in technological progress. (Atlantic Council, 2026; IFRI, 2018)
The United States leads in creating advanced AI models and investing in data centers, while China has quickly expanded its infrastructure and is exporting energy and digital solutions through coordinated state efforts. Despite Europe’s strong regulatory framework—embodied by the AI Act—it still relies on U.S. hyperscalers for cloud and AI workloads and faces capacity and investment constraints. (Bruegel, 2025; Carnegie Endowment, 2025).
But the issue goes beyond technology. Control over AI infrastructure impacts economic competitiveness, military strength, intelligence capabilities, and the ability to set global standards. Countries and regions that do not develop or acquire cutting-edge AI technologies risk economic stagnation and reduced influence. For Europe, the key question isn’t whether to pursue strategic autonomy in AI and data infrastructure but how to do so in a world where the U.S. and China follow different rules.
This paper will analyze the geopolitics of AI infrastructure and suggest strategies for Europe to achieve strategic autonomy—ambitious yet feasible. It will propose ways for Europe to develop sustainable infrastructure, maintain sovereignty in critical areas, and show selective leadership through a framework nation strategy.
The New Geopolitics of Compute
AI computing infrastructure has become a strategic asset, similar to ports, oil pipelines, semiconductor manufacturing facilities, and power grids during times of geopolitical competition. Training and operating advanced AI models require enormous amounts of power, specialized hardware like GPUs, extremely high-speed Internet access, secure locations, and effective cooling systems. Increasingly, ownership of these resources will determine which countries, companies, or other actors can develop and profit from advanced AI technology. (Standford HAI, 2026)
The scope of this development effort is unprecedented. In 2024 alone, the United States is building 5.8 GW of data center capacity, while the European Union is deploying an additional 1.6 GW. China’s demand for electricity used by data centers is expected to reach 277 TWh annually by 2030—twice the level of 2023. It’s no longer just about engineering feats; it’s about managing enormous amounts of power. The infrastructure race is closely tied to the broader competition between the U.S. and China. Export restrictions on advanced chips imposed by the U.S. aim to hinder China’s progress in innovative AI technologies.
China is striving to develop its own chip production, supply chains, energy generation, and data centers. Other countries and regions must decide whether to adopt U.S.-developed technology, which offers security guarantees, or to rely on China’s infrastructure, which currently lacks political strings but may create future dependence. Energy now plays a crucial role, as AI data centers demand substantial energy resources. One large center can consume as much electricity as a small city. This leads to new dependencies on energy infrastructure, production, and grid stability. Countries capable of producing affordable, reliable energy will have a competitive edge in attracting AI investments. (OECD, 2025)
American Advantages and Structural Vulnerabilities
Currently, the U.S. holds a significant edge in developing innovative AI systems and private sector investments in data centers. American companies have secured a dominant position in the global cloud computing market and are investing hundreds of billions of dollars in building new data centers. These trends, along with a strong venture capital system, advanced research organizations, and a culture of continuous improvement, position the country to outperform competitors in innovation and implementation.
However, this advantage comes with some weaknesses. Energy constraints, for instance, pose a major challenge for expanding the U.S. data center network. Permit delays, power grid limitations, local opposition, and outdated transmission infrastructure complicate construction efforts. Additionally, the U.S. still depends on semiconductor manufacturers from Taiwan, making it vulnerable to supply disruptions. Furthermore, the heavy concentration of AI capabilities among a few tech giants presents significant risks.
American policies, such as the CHIPS and Science Act, export control measures, and diplomatic efforts, aim to secure allied supply chains. However, these policies could strain relationships with European allies, as they risk entanglement in great-power competition or force choices among competing technologies.
The U.S. appears to be reconsidering the sustainability of its model. Private investments in AI infrastructure are driven by high profits, but energy limitations and supply chain issues could hinder growth. Intense global competition is likely to compel not only the U.S. but also China to invest more heavily in achieving technological self-sufficiency.
Currently, the U.S. holds a significant edge in developing innovative AI systems and private sector investments in data centers. American companies have secured a dominant position in the global cloud computing market and are investing hundreds of billions of dollars in building new data centers. These trends, along with a strong venture capital system, advanced research organizations, and a culture of continuous improvement, position the country to outperform competitors in innovation and implementation.
However, this advantage comes with some weaknesses. Energy constraints, for instance, pose a major challenge for expanding the U.S. data center network. Permit delays, power grid limitations, local opposition, and outdated transmission infrastructure complicate construction efforts. Additionally, the U.S. still depends on semiconductor manufacturers from Taiwan, making it vulnerable to supply disruptions. Furthermore, the heavy concentration of AI capabilities among a few tech giants presents significant risks.
American policies, such as the CHIPS and Science Act, export control measures, and diplomatic efforts, aim to secure allied supply chains. However, these policies could strain relationships with European allies, as they risk entanglement in great-power competition or force choices among competing technologies.
The U.S. appears to be reconsidering the sustainability of its model. Private investments in AI infrastructure are driven by high profits, but energy limitations and supply chain issues could hinder growth. Intense global competition is likely to compel not only the U.S. but also China to invest more heavily in achieving technological self-sufficiency.
China’s State-Driven Infrastructure Surge
Unlike the US approach, the Chinese method appears quite different. The Chinese government actively coordinates investments and planning related to energy and digital infrastructure. It also promotes national champions in artificial intelligence and related fields. For example, the “Eastern Data, Western Computing” project enables China to build large data centers in its western regions, which are then connected to consumption centers in the east. (Strider Technologies, 2025)
Undoubtedly, there are several advantages to China’s approach. On one hand, the government can coordinate permitting processes, energy supply, land use, and industry policies more swiftly to meet its goals. On the other hand, China exports its strategy through the Belt and Road digital infrastructure projects, offering partner countries energy systems, data centers, and connectivity under fewer regulations.
However, China also faces a new set of challenges. Due to US export controls, China cannot access the most advanced chips and must rely on domestically produced, less efficient alternatives. The rapid increase in energy consumption is notable; despite being a leader in renewable energy, integrating these sources to support AI remains difficult. Additionally, there is growing geopolitical resistance to Chinese technology in Europe, North America, and parts of Asia.
There are potential issues with over-investment in China. State-led investments may lead to misallocations and projects that are economically unjustified but driven by political motives. Ultimately, as the technology matures, it will be necessary to reevaluate the infrastructure investments.
Europe’s Strategic Autonomy Dilemma
The structural disadvantage of Europe is clear. Europe has excellent research facilities, regulation, and industry in specific sectors. However, when it comes to innovation in AI frontier areas and building the necessary infrastructure, Europe lags behind not only the USA but also China. European private and government institutions heavily depend on American cloud providers (AWS, Microsoft Azure, Google Cloud), creating a reliance on US technology companies for their AI efforts. (Bruegel, 2025)
A significant issue is energy, which is especially restrictive. Running a data center requires a reliable, green, and affordable power source. Many European countries face delays in building data centers due to complex licensing, inadequate grid infrastructure, local protests, and competition for renewable resources from other industries. Despite robust data protection and security regulations, Europe’s regulatory environment has been criticized for limiting innovation and investment compared to the more lenient frameworks in the United States and China.
However, technological sovereignty has become a key concern for Europe, gaining political importance, especially in France and within EU institutions. Initiatives include the AI Continent Action Plan, AI gigafactories, and the Cloud and AI Development Act. Still, turning these ambitions into practical industrial and infrastructure capabilities is challenging. Europe lacks both the level of private investment seen in the US and the coordinated state effort demonstrated by China.
This creates a critical dilemma for Europe. It cannot replicate the American model based on private and venture capital investments, nor can it adopt China’s approach without altering its political and economic systems. On the other hand, it must find its own way to leverage regulatory power and develop its industrial base.
Pathways for European Leadership
Despite the challenges listed above, there are still certain areas where Europe can leverage its advantages.
First, Europe may be particularly well-positioned to develop sustainable, energy-efficient data centers and cloud services. With its experience and knowledge of energy production and energy system engineering, along with high levels of industrial efficiency, European companies might create unique solutions in sustainable computing.
Second, Europe could focus on developing secure infrastructures of cloud and AI technologies tailored to specific uses. Even if it will be rather difficult for Europe to outpace American hyperscalers in the commercial use of these technologies, it is still possible to develop infrastructure within Europe that will be useful to government, the military, critical infrastructure, and regulated industries. (European Comission, 2025)
Third, another idea, using “framework nations,” which European countries have already used for defense and capability development, could also prove successful for technological infrastructure development. Nations with specific capabilities could be leaders in different sectors while coordinating actions and requirements with their allies. Thus, for example, France, with its capabilities in nuclear energy and supercomputing, would be an excellent candidate for framework nations in AI infrastructure with high energy use. Meanwhile, the Netherlands and Nordic countries excel at designing energy-efficient data centers. Finally, Germany has the necessary capabilities to use AI technology in industry. (Fiott, 2026)
The decision-making process may prove extremely difficult. It was impossible for Europe to catch up with the US and China in terms of the development of the AI infrastructure. Europe will need to choose the sectors in which it can achieve interdependence and autonomy.
Thus, an analysis of the EU’s weaknesses will be needed.None of the nations belonging to the organization can move forward at the same pace or with the same zeal. Perhaps it may be more rewarding to form an alliance of those who wish and grow independently with their flexible systems. (OECD, 2025)
Bibliography
European Commission (2025) “AI Continent Action Plan” and related initiatives. Available at: https://digital-strategy.ec.europa.eu/en/policies/ai-continent-action-plan
Atlantic Council (2026) “Digital Sovereignty: Europe’s Declaration of Independence?”. Available at: https://www.atlanticcouncil.org/in-depth-research-reports/report/digital-sovereignty-europes-declaration-of-independence/
Carnegie Endowment for International Peace (2025) “The EU’s AI Power Play: Between Deregulation and Innovation”. Available at: https://carnegieendowment.org/russia-eurasia/research/2025/05/the-eus-ai-power-play-between-deregulation-and-innovation
Bruegel (2025) “How to accelerate investment in sovereign AI computing infrastructure, models and services in the EU?”. Available at: https://www.bruegel.org/event/how-accelerate-investment-sovereign-ai-computing-infrastructure-models-and-services-eu
IFRI (2026) Reports and analysis on European Technological Sovereignty. Institut Français des Relations Internationales. Available at: https://www.ifri.org/en/thematiques/technology/technological-competition-and-digital-sovereignty/european-technological
Stanford HAI (2026) “The 2026 AI Index Report”. Available at: https://hai.stanford.edu/ai-index/2026-ai-index-report
Brookings Institution (2026) “How will the United States and China power the AI race?”. Available at: https://www.brookings.edu/articles/how-will-the-united-states-and-china-power-the-ai-race/
OECD (2025) “Competition in artificial intelligence infrastructure”. Available at: https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/11/competition-in-artificial-intelligence-infrastructure_69319aee/623d1874-en.pdf
Special Competitive Studies Project (2025) “Gaps Analysis 2025 Report”. Available at: https://www.scsp.ai/wp-content/uploads/2025/01/Gaps-Analysis-2025-Report.pdf
Strider Technologies (2025) “China’s AI Infrastructure Surge”. Available at: https://content.striderintel.com/wp-content/uploads/2025/05/Strider-SCSP-China-AI-Infrastructure-Surge-Report.pdf
About Atuf Guliyev
Atuf Guliyev is a specialist in international relations, with extensive interest in diplomacy, and global governance. He is currently pursuing studies in Diplomacy and Global Governance at the Brussels School of Governance (Vesalius College, Vrije Universiteit Brussel) and is developing a doctoral research project examining the selective Europeanisation of French foreign policy since 2010 and its consequences for European strategic autonomy. His academic interests include French foreign and security policy, the EU's Common Security and Defence Policy (CSDP), norm entrepreneurship in European crisis management, and the nexus between security and humanitarian action in the Mediterranean and Sahel regions.
View all posts by Atuf Guliyev →
No comments:
Post a Comment