A new era of self-reliance: Navigating technology sovereignty

Greater Europe is leading the drive

In September 2024, the European Union released the “Draghi report,” which outlined recommendations for improving overall European economic competitiveness.⁶ Part of the report focused on how to potentially advance its domestic tech sector and how the sector could improve innovation, technology adoption, and worker productivity. The report preceded the launch of the EuroStack Initiative—a call from over 200 European companies and organizations for “radical action” around increasing technology sovereignty.⁷ This included advocating for buying European, pooling and leveraging existing assets more effectively, focusing less on research and development and more on productization, ensuring adequate capital, and protecting data for European cloud users. Overall efforts of the European Commission are being led by a designated Commissioner for Technology Sovereignty. This continues a long history of the European Union seeking sovereignty in tech, believing that sovereign solutions are best suited for supporting the EU philosophy, values, and principles—embodied in frameworks such as the General Data Protection Regulation, the Digital Services Act, and the AI Act.

Initial fervor and expectations may have moderated somewhat since early 2025, as reflected in the recent EU International Digital Strategy, which is focused more on cooperation with other countries around AI, semiconductors, quantum computing, and cybersecurity.⁸ The debate on the best strategic approach to take is ongoing, but there’s likely to be over €100 billion in public and private investment for European cloud computing, AI data centers and companies, semiconductors, and satellite communications efforts over the next five years.

Cloud computing

Local European cloud providers comprise a very small percentage (less than 20%) of the overall market.⁹ They would require significant investment and time to develop into true competition for global hyperscalers. What’s more likely to happen is that global players will increasingly provide European-specific adaptations of their capabilities. Amazon Web Services (AWS) announced that it will invest almost €8 billion in a European Sovereign Cloud located in Germany. Goals for the project include allowing customers to keep their data in the European Union, providing independence, and ensuring it is led, operated, secured, and governed by EU citizens.¹⁰ Microsoft has also announced a set of commitments to Europe—specifically around AI, cybersecurity, privacy, resiliency, and economic competitiveness—and a Microsoft Sovereign Cloud platform and solutions.¹¹

AI models and data centers

There are several initiatives from both the government and commercial sectors looking to improve overall AI capabilities. The European Commission’s AI Continent Action Plan seeks to develop a series of AI factories and gigafactories across Europe, building on existing supercomputing infrastructure, and driving net-new investment through the InvestAI program.¹² This program will make €20 billion available for up to five new AI gigafactories that will enable the creation of advanced, cutting-edge AI models known as “sovereign frontier models.” The Action Plan also looks to improve data availability for AI models, the use of AI applications, and skills and workforce development. On the commercial side, NVIDIA and Perplexity are teaming up to help train and make open-source, localized AI models widely available.¹³ NVIDIA is a backer, along with investment firm MGX, Mistral AI, and others, to create Europe’s largest AI data center by 2028 at a cost of €8.5 billion.¹⁴ There is also Stargate UK, a phased effort to build out AI infrastructure across the country and accelerate domestic AI adoption.¹⁵

Semiconductors

Much like the United States, Europe wants to onshore more semiconductor manufacturing, strengthen the resilience of its supply chains, advance a stronger local ecosystem, and boost European companies. To that end, the EU Chips Act (2023) established a fund, pilot lines for experimentation, a collaborative design platform, and competency centers, and provides resources for quantum chips—€43 billion in total investment through 2030.¹⁶ There is already significant commercial investment happening, including a FinFET (field-effect transistor) pure-play foundry, a “Smart Power Fab,” and a silicon carbide chip manufacturing plant, among others.¹⁷

Satellite communications

Another key initiative for Europe is building its own satellite communications constellations to reduce dependence on providers outside the bloc—ensuring secure and reliable services for military, government, and commercial applications. The two main efforts consist of the Infrastructure for Resilience, Interconnectivity and Security by Satellite (IRIS²) constellation and Eutelsat OneWeb. IRIS² will eventually consist of almost 300 satellites in multiple orbits at a cost of about €11 billion.¹⁸ Eutelsat is looking to accelerate its efforts to build out and enhance its OneWeb low earth orbit satellite internet constellation, which currently has more than 630 satellites in orbit.¹⁹ It recently received fresh investment from both the UK and French governments to make this happen.²⁰ In an increasingly crowded and competitive market, it will take time (IRIS² completion is planned for 2031) and significantly more investment for both of these constellations to reach the point where they can effectively challenge current services and fully support European needs.²¹

What about the rest of the world?

Although Europe is driving a significant amount of technology sovereignty activity, other countries and geographic regions are pursuing their own unique and innovative approaches, with most of the efforts focused on AI. This isn’t meant to be comprehensive but rather to show the breadth and depth of global activity.

• South Korea: South Korea aims to develop sovereign AI capabilities based on its language and tailored to its culture.²² One example is Kakao partnering with OpenAI on new personalized digital services.²³ To bolster domestic infrastructure, SK Group and AWS announced that they will jointly build South Korea’s largest AI data center by 2029, at an estimated cost of US$5 billion.²⁴
• Japan: The country is looking to advance its AI capabilities and reinvent its domestic semiconductor industry through the Rapidus initiative—a new company focused on 2-nanometer technology—and a proposed US$65 billion government investment package through 2030.²⁵
• Africa: Africa’s first AI factory, powered by NVIDIA’s AI and accelerated computing platform capabilities, will be located in Cassava Technologies’ data center facilities in South Africa—with plans to expand to other locations across the continent, including Egypt, Kenya, Morocco, and Nigeria.²⁶
• India: There is a strong drive for self-reliance across all layers of the tech stack in India, and government programs like the India Semiconductor Mission and IndiaAI are working to address those needs.²⁷ India will face some unique challenges in developing its AI models, including compute availability, multiple languages to support, and a lack of high-quality training data.²⁸ India’s strong domestic digital capabilities developed for the India Stack could be expanded and exported to other countries to create a new, competitive digital ecosystem.²⁹
• Canada: The Government of Canada’s sovereign AI compute strategy focuses on improving private investment, public infrastructure, and funding access to compute resources.³⁰ It has also announced a partnership with domestic AI firm Cohere to explore how they can both improve Canada’s overall technological capabilities.³¹ In addition, several Canadian telecommunications companies are planning to build sovereign AI data centers, including TELUS, SaskTel, and Bell.³²
• Middle East: Many countries in the region are increasing their investment in sovereign cloud and AI data centers—including major projects like the Stargate UAE initiative, a 1 gigawatt AI cluster.³³ The Public Investment Fund of Saudi Arabia established HUMAIN in 2025, a new company looking to develop end-to-end AI infrastructure with the help of global partners like AWS, NVIDIA, and others.³⁴ US$23 billion in investment has been announced in relation to these partnerships.³⁵

Dealing with the consequences

What happens if, as expected, most governments pursue robust technology sovereignty policies and programs in the near future? There are a variety of potential benefits to having greater control over end-to-end technological capabilities. These include economic ones such as greater tax revenue and private capital investment, better employment opportunities for citizens, and a greater chance for homegrown tech companies to flourish. By being more self-reliant, there is a belief that overall resiliency can be improved, privacy and security can be enhanced, and exposure to potential political disruption from foreign countries can be reduced. Additionally, when it comes to AI, if foundational models are created within a country, they can better reflect its local language, customs, and data sets.

We could also see challenges arise, such as:

• Shifting investment flows. Foreign direct investment, mergers and acquisitions, and joint ventures could potentially face increasing numbers of conditions and requirements. Venture capital investment could also shift focus. Will venture capital firms put national strategic interests above more global opportunities?³⁶
• Increased fragmentation. Taking a more insular, zero-sum approach may lead to lower levels of collaboration, fractured international relationships, and fewer academic partnerships. We could also see reduced cross-border flows of data and proprietary communications infrastructure, as well as an increase in the number of standards and regulations.
• Workforce impacts. With countries putting greater emphasis on domestic technological capabilities, overall global mobility for highly skilled workers could shift. This could be especially acute in critical areas like AI, cybersecurity, and chip design. There will also likely be greater investment to bolster broader national workforce capabilities.³⁷
• Environmental impacts. A large increase in the construction of fabs, labs, data centers, and associated supporting infrastructure will put strains on resources. Some countries’ power grids are already maxed out, and with new data centers demanding thousands of additional megawatts, they could compete with residential energy needs.³⁸ There is also the challenge of utilizing non-polluting, low- or zero-carbon electricity sources.³⁹
• New partnerships. Not everyone can do it alone. In the future, we may see more bilateral agreements, regional frameworks, and nontraditional technology alliances seeking to capitalize on each other’s strengths.⁴⁰
• Overcapacity. How many foundational AI models can the market support? There are massive global capital expenditures happening just for AI infrastructure—estimated to be almost US$3 trillion through 2028. Will all of it produce a return on investment?⁴¹ Long-term demand may not meet extraordinary expectations, and new technological innovations may lessen the need for current approaches.⁴²

The Bottom Line: Prepare for a more self-reliant future

In 2026, expect the drive for technology sovereignty to continue with more debate, government action, and investment activity. While the motivations and eventual outcomes of this drive are open to discussion, action is underway—and more will be taken—because many believe that the future prosperity of their countries and regional blocs is at stake.

• Audit global dependencies. Identify and assess all critical dependencies—data flows, public cloud, vendors, supply chains, financial, and regulatory. Build new, and strengthen existing, partnerships that could provide the most global flexibility. Be able to transparently explain your global operations.
• Anticipate regulatory complexity. Prepare for a fast-evolving regulatory environment. Expect new rules on data localization, cybersecurity, mergers and acquisitions, and capital flows. Pinpoint where your business is most exposed and build scenario plans now. Bolster your compliance programs.
• Revisit your cloud strategy. Think about the balance between your public and private cloud capabilities. Adopt a multicloud or sovereign cloud model to enhance resilience and compliance. Prioritize portability, interoperability, and control across environments. Ensure your vendors support automatic compliance for data storage, processing, and transfer. Prepare contingency plans to stay agile in the face of geopolitical shifts.
• Strengthen talent resilience. Know where your critical talent comes from—and what happens if access is disrupted. Develop alternative talent-sourcing strategies and leverage government workforce programs and university partnerships to grow specialized skills.