Europe's tech sovereignty push: What should businesses expect

Control over AI becomes existential for European leaders 

European tech sovereignty agendas are moving rapidly from political rhetoric to concrete policy, with AI at its centre. Governments increasingly view AI not only as a driver of productivity and economic growth, but also as a strategic capability linked to national security. This, in turn, drives their concerns about Europe’s reliance on foreign providers across the AI stack – from advanced semiconductors and compute capacity to frontier models and applications. 

Framing tech sovereignty in existential terms is a response to both security and economic dynamics. Growing geopolitical tensions, concerns about the reliability of the US as a long-term strategic partner under the second Trump administration, the weaponisation of supply chains, and the growing importance of AI for modern warfare and national security have reinforced calls for greater control over this transformative technology.  

European governments are also determined not to repeat the mistakes of the internet era and to ensure that a greater share of the value generated by the AI transformation remains in Europe to help revive the continent’s sluggish economic growth, innovation, and productivity, particularly compared to the US and China.  

The result is not a single, coherent policy agenda or a uniform political trend. Rather, governments hold diverging positions that result in a spectrum of approaches. This reflects a broader tension at the heart of Europe’s AI debate: while policymakers increasingly accept the case for accelerating AI adoption to remain competitive, they also want a greater share of the value and strategic control associated with AI to remain in Europe. In an EU context, this ties into the EuroStack initiative’s thinking, which argues that Europe should develop greater control over the critical layers of its technology stack, from infrastructure to platforms and applications.  

As a result, approaches range from relatively open sovereignty models, focused on resilience and support for domestic ecosystems, to more interventionist frameworks involving European preference requirements and restrictions on foreign providers. This proliferation of responses across the EU and the UK carries significant implications for businesses and investors operating across the AI ecosystem. 

The EU’s ambitious agenda 

Debates on EU “strategic autonomy” and tech sovereignty are not new. However, the growing EU-US divide since the start of the second Trump administration has led the EU to accelerate its efforts to reduce reliance on US technology, including by favouring local economic actors. EU policymakers also increasingly fear that failure to promote home-grown AI development and adoption could deepen the innovation and competitiveness gap with the US and China identified by Mario Draghi. The European Commission is responding to this new urgency through multiple policy instruments, centred on the upcoming Tech Sovereignty Package, which aims to boost EU sovereignty across the different layers of the AI stack. 

The Cloud and AI Development Act (CAIDA) – likely the most consequential component of the Tech Sovereignty Package – will seek to triple EU data centre capacity by 2030 by designating "industrial valleys" offering favourable permitting and grid access for developers. The Chips Act 2 will aim to strengthen Europe’s semiconductor ecosystem by supporting demand, consolidating the link between R&D and industrial deployment, and facilitating financing for new chip fabs. In parallel, the Commission is planning to launch a call for tenders to develop AI Gigafactories – public-private AI infrastructure hubs to train and run advanced AI systems – although it is unclear whether sufficient public funding will be mobilised. It remains uncertain whether these efforts will actually reduce Europe’s reliance on foreign providers, particularly given the dominance of US firms across the most advanced parts of the AI stack. 

The Commission will also likely explore explicit sovereignty requirements and “European preference” criteria to restrict market access for foreign AI and cloud providers. CAIDA will likely contain sovereignty requirements in the procurement of cloud services based on a risk-tiered approach. Key details remain under discussion, including the definition of sovereign cloud, and the strictness of requirements to be imposed on cloud service provision to public administrations – and, potentially, to private companies in critical sectors. Similarly, the forthcoming Public Procurement Act may introduce “Made in Europe” criteria for procurement in certain strategic sectors, while the proposed Cybersecurity Act revision could restrict access to critical infrastructure contracts for any vendor established in a designated “third country posing cybersecurity risks”. 

This EU sovereignty agenda may appear coherent and comprehensive, but it remains hindered by significant disagreements within the Commission and among Member States. There are unresolved differences on the meaning of tech sovereignty, the level of ambition required from legislation, and the extent to which European companies should be favoured over foreign providers. At its core, the debate reflects broader disagreements between France and Germany. While the former favours strict European preference criteria, Germany’s export-driven economy is vulnerable to retaliation, leading Berlin to favour a more open conception of sovereignty that need not exclude US providers outside critical sectors. Ultimately, much will depend on how the EU defines the scope and obligations attached to sovereignty requirements.  

However, the two countries have been working towards a common definition of tech sovereignty, which the French Government is now consulting on. Convergence between the EU’s two major countries would give them significant influence on the ultimate level of ambition of the EU’s sovereignty agenda. Either way, businesses should prepare for material commercial and market access implications from these processes which are currently taking shape. 

The UK’s middle path 

The debate around tech sovereignty has also intensified in the UK in recent months. Rising concerns about dependence on global technology supply chains and intensifying political scrutiny of UK dependence on large US technology firms are piling pressure on ministers to articulate a clearer strategy for strengthening domestic capability and resilience in critical technologies, such as AI.  

However, the UK Government is charting a more modest path compared to the EU, focusing on three ambitions: building leverage in the AI technology stack, accelerating adoption across the economy, and retaining sovereign control over a narrower set of capabilities – compute, sensitive data, and resilience. 

This approach was crystallised in a major speech by Technology Secretary Liz Kendall at the end of April. Framing AI through the dual lens of geopolitical rivalry between the US and China and rapid technological disruption, Kendall argued that the UK should reduce critical dependencies while increasing resilience in areas of national strategic priority. In practice, this has translated into an effort to build on the UK’s strengths by directing government support to promising domestic start-ups, while working with like-minded international partners and fellow “middle powers” in areas such as AI security standards and joint investment in tech priorities. 

The clearest manifestation of this approach has been the launch of Sovereign AI. This sovereign venture fund, backed by £500m, will deploy funding and targeted compute, talent, and regulatory support to domestic AI companies, including supercomputing access, fast-track visas, and procurement opportunities, signalling a greater willingness to use government demand to support domestic firms.  

However, this tech sovereignty agenda continues to be hindered by scale and structural constraints. Public funding remains relatively small compared to the scale available in the US, while issues around high energy costs, limited data centre capacity, and late-stage capital gaps remain unresolved.  

For now, the immediate commercial impact for businesses remains more limited than in the EU, although a change in Prime Minister may bring a different approach. In particular, commercial risks may emerge for businesses if the government concludes that certain services supporting public administration or critical infrastructure need to be brought under sovereign control. 

Trade and geopolitical repercussions are likely  

Albeit in its infancy, the European tech sovereignty agenda has already been a source of concern among trade partners. The US has long been sensitive to EU non-tariff barriers in the tech sector, but the Trump administration has been highlighting a much wider set of barriers and has shown a greater willingness to retaliate (see our blogpost on US tariff policy here). US Ambassador to the EU Andrew Puzder has explicitly warned the Commission against introducing protectionist measures, suggesting the upcoming Tech Sovereignty Package may compromise the EU-US trade deal clinched in Scotland last summer.  

China, while more cautious publicly, has similarly voiced concerns about the direction of travel of EU policy. Beijing is particularly concerned by the proposed revision of the Cybersecurity Act, which would give the Commission effective powers to restrict Chinese technology providers’ access to European critical infrastructure markets. 

“European preference” criteria are likely to become a source of friction even with countries whose trade relations with the EU are less strained. The Commission’s proposal to introduce “Made in Europe” requirements in the Industrial Accelerator Act (see our blogpost here) already offers an indication of the tensions to come, with allies like the UK and Japan lobbying the Commission and Member States to be included under European preference frameworks. 

What this means for business 

Europe’s shift towards tech sovereignty will carry important risks and implications for companies and investors, though the impact will vary sharply by sector, size, and sensitivity. Hyperscalers and foreign companies operating in critical sectors like defence and public services will be most exposed and could face growing market access barriers, as EU institutions and national governments increasingly prioritise European providers in their public procurement, investment frameworks, and regulatory compliance schemes. Other AI, software, and consumer-facing firms may also face higher compliance costs through new certification regimes, data localisation rules, and sovereignty-based standards, as well as growing customer demand for sovereignty-aligned offerings. 

European start-ups and incumbents in compute, models, and chips stand to gain from public investment, access to public compute, and preferential procurement. However, these opportunities may come with a risk of retaliation from trade partners and possible market access barriers in third countries, as well as restricted or degraded access to foreign tools. 

Across the board, companies will face mounting pressure to demonstrate how they contribute to Europe’s sovereignty objectives, whether through investment in domestic R&D, partnerships with European firms, localisation of infrastructure, or support for broader resilience and security goals. At the same time, sovereignty regimes are likely to come into growing conflict – between Europe and the US, between Member States, and between the UK and the EU.  

This will require a tailored, locally grounded approach to government engagement. A single corporate narrative will not work simultaneously in Brussels, London, Paris, and Berlin. Companies operating across markets that are increasingly at odds on industrial and trade policy will therefore need to carefully calibrate their global business models and engagement strategies to local political expectations. 

This blog was written by Manager Giacomo Borsetti, Associate Director Orla Middlemiss, and Manager Léna Rigoulay, with additional input from Associate Directors Euan Edwards and Miro Guzzini, and Partner Benoît Roussel. 

This is the first in a series on Europe’s tech sovereignty agenda. The next piece will examine the Tech Sovereignty Package once published. For our earlier work on AI sovereignty in the Asia-Pacific region, see Sovereign AI in APAC: what it means for business. 

Flint Global has exceptional people in Brussels, London, Paris, Berlin and across Europe and Asia-Pacific, able to advise tech companies and investors on navigating Europe’s evolving sovereignty agenda. To discuss the implications for your business, get in touch.