Europe’s AI Moment: Relevance Without Capability

Europe is attempting to remain strategically relevant between the United States and China in artificial intelligence. It has world-class researchers, a large internal market, substantial savings, established industrial expertise, and companies developing commercially viable AI technologies. Yet it has not demonstrated the ability to bring these resources together at sufficient scale to shape AI development and deployment at the frontier.

Whether European policymakers ultimately seek parity with the United States or China is uncertain. What is clearer is that the European Union wants to avoid strategic marginalization. Its immediate objectives are to protect its market, preserve industrial and security autonomy, reduce critical dependencies, and retain influence over technologies that will increasingly determine economic and geopolitical power.

The challenge is not simply to raise more capital or build more computing infrastructure. It is to convert Europe's existing resources into a coherent, competitive industrial ecosystem. That requires effective capital markets, coordinated infrastructure, commercially successful companies, predictable regulation, and institutions capable of executing complex programs.

Europe has many of the necessary components. What remains uncertain is whether it can make them work together.

The capital paradox

In an interview with China Media Group, published by ShanghaiEye on November 2, 2026, European Central Bank President Christine Lagarde reiterated concerns about the direction of European savings and the financing of artificial intelligence. Her remarks followed her September 14 speech in Vienna, A New Age of Capital: Growth, Sovereignty and AI, in which she argued that European savings are once again helping finance transformative technologies developed primarily in the United States (Lagarde, 2026; ShanghaiEye, 2026).

Lagarde identified a historical parallel. During the nineteenth century, European savings financed major American infrastructure projects, including railways. Much of the resulting economic growth and industrial capability remained in the United States. She sees a similar risk today: Euro area households hold approximately €440 billion in US technology companies, while European businesses struggle to access the equity financing needed to develop and scale their own technologies.

Her argument is not that European savings should never flow abroad. Capital will always seek returns across borders. The problem is that Europe has not developed sufficiently effective financial markets to channel a greater proportion of its savings into productive investment at home. European households save approximately €1.4 trillion annually, yet European technology companies frequently face a more difficult financing environment than their American counterparts (Lagarde, 2026).

The implications extend beyond investment returns. European savings may help finance the technology ecosystem on which European businesses increasingly depend. European companies may then purchase foreign models, cloud services, computing capacity, and other critical infrastructure rather than develop or control those capabilities themselves.

This creates a strategic imbalance. Europe supplies capital, provides a large market, and establishes regulatory conditions, but captures a comparatively limited share of the companies, infrastructure, and intellectual property generating the value.

AI makes this imbalance particularly consequential. It is becoming a general-purpose industrial capability, influencing manufacturing, healthcare, defense, scientific research, cybersecurity, public administration, and almost every sector that relies on software and data. The ability to develop, deploy, and integrate AI will increasingly affect productivity, industrial competitiveness, and national security.

Europe cannot simply assume that it can purchase every capability it needs indefinitely. Foreign technology will remain an important part of its ecosystem, but dependence becomes strategically consequential when critical infrastructure, essential services, and industrial systems rely on technologies whose development, ownership, and availability Europe does not control.

The question is therefore not whether European capital should remain inside Europe. It is whether Europe can create an investment environment in which competitive European companies become attractive destinations for that capital.

The scale of the gap

The investment imbalance is substantial.

In 2025, AI companies attracted 61% of global venture capital investment, equivalent to $258.7 billion of a $427.1 billion total. Companies based in the United States received approximately 75% of global AI venture capital deal value, compared with 6% for EU27 companies. US investors also accounted for approximately 56% of outgoing global AI venture capital investment, against 7% for EU27 investors (OECD, 2026).

These figures measure venture capital rather than total AI investment, but they reveal a significant imbalance in the financing available to companies seeking to commercialize and scale AI technologies. They also show the growing importance of large financing rounds. Deals exceeding $100 million accounted for approximately 73% of global AI venture capital investment value in 2025 (OECD, 2026).

The concentration matters. Developing competitive AI companies requires more than funding research or supporting early-stage startups. Companies need access to successive rounds of financing, infrastructure, energy, specialized talent, commercial customers, and international markets. A funding environment that supports promising research but cannot finance the transition to industrial scale will struggle to produce globally competitive businesses.

The European Court of Auditors identified related weaknesses in its 2024 assessment of the EU's AI ambitions. It found that European and national measures were not effectively coordinated, that the Commission lacked sufficient governance tools and information, and that the initiatives had not accelerated AI investment in line with global leaders.

The auditors recommended stronger coordination, improved monitoring, more focused investment, and greater support for commercializing research within Europe (European Court of Auditors, 2024).

The underlying problem is not a shortage of European research or industrial expertise. It is the difficulty of connecting research, finance, infrastructure, commercialization, and market demand into a functioning system.

What the EU is doing

The European Commission has begun to respond through the AI Continent Action Plan, AI Factories, AI Gigafactories, InvestAI, and the AI Act (European Commission, 2025a, 2025b; European Parliament and Council of the European Union, 2024).

The AI Continent Action Plan represents a shift toward a broader industrial strategy. Its objectives include expanding computing infrastructure, improving access to data, supporting AI adoption in strategic industries, developing skills, and simplifying regulatory implementation.

AI Factories are intended to provide startups, researchers, and established companies with access to computing resources and technical support. AI Gigafactories are intended to provide substantially greater computing capacity for training and developing more complex AI models.

The Commission has identified approximately €10 billion in combined investment by the Commission, Member States, and associated countries for supercomputing infrastructure and AI Factories over the 2021–2027 period. InvestAI includes a €20 billion initiative intended to mobilize financing for up to five AI Gigafactories. The broader InvestAI initiative aims to mobilize €200 billion in AI investment across Europe (European Commission, 2025a, 2025b, n.d.).

These are significant commitments. They demonstrate that the EU recognizes the relationship between computing infrastructure, capital, industrial development, and technological competitiveness.

However, the figures describe different funding and investment ambitions. They should not be treated as equivalent amounts of public money already committed or deployed. The €200 billion objective, in particular, represents investment to be mobilized rather than a completed public expenditure.

This distinction matters because Europe has repeatedly demonstrated an ability to announce ambitious programs. The more difficult question is whether it can translate those commitments into operational infrastructure, successful companies, commercially viable products, and measurable industrial outcomes.

The gap between ambition and execution

The fairest assessment of the European response is that it is necessary, incomplete, and not yet supported by a convincing record of execution.

It is necessary because Europe cannot protect its industrial and security interests through regulation alone. It needs access to capital, compute, energy, chips, data, specialized talent, and commercial infrastructure.

It is incomplete because the initiatives announced so far do not resolve the structural weakness of European growth financing. Early-stage research support and computing infrastructure are important, but companies also need substantial, patient capital to expand operations, enter international markets, and compete with established technology businesses.

It is not yet supported by a convincing record of execution because the EU has struggled to demonstrate that its institutional mechanisms can consistently turn ambitious programs into measurable industrial outcomes.

This is not an argument against European investment or industrial policy. It is an argument for measuring performance rather than relying on commitments.

A useful framework should distinguish five stages:

Each stage requires independent measurement. Announced investment is not deployed capital. Deployed capital is not necessarily delivered capability. Operational infrastructure does not automatically produce competitive companies or economic returns.

This framework should become the standard by which European AI initiatives are evaluated.

The EIC Fund: a limited success

The European Innovation Council Fund provides an important precedent for European direct equity investment.

It has demonstrated that the EU can invest directly in deep-technology companies and attract additional private capital. By 2025, the EIC Fund reported more than €1 billion in approved investments in 272 companies and more than €2.6 billion in additional investment mobilized since its launch in 2020 (European Innovation Council and SMEs Executive Agency, 2025).

These are meaningful achievements. They demonstrate that European institutions can participate in equity financing and help attract private investors to technology businesses.

However, the EIC Fund should be regarded as a limited success rather than a proven model for financing Europe's next generation of strategic AI companies.

Its early operation was marked by significant delays, fragmented governance, disputes over investment authority, limited internal investment capacity, and reluctance to lead financing rounds. The European Court of Auditors documented the slow initial deployment of EIC equity investments, including the fact that, by the end of 2022, only 2% of the EIC Fund’s Horizon Europe budget for 2021–2022 had been disbursed (European Court of Auditors, 2024, para. 63 and Table 1 in para. 62).

The Fund subsequently increased its investment activity. Its reported investment and co-investment figures are evidence of progress, but they do not, by themselves, establish that its original institutional weaknesses have been resolved or that it can provide the scale and speed required for frontier AI investment.

The absence of sufficiently timely, independent follow-up reporting on those structural issues makes it difficult to establish whether the Fund has achieved sustained institutional improvement. Positive portfolio figures cannot substitute for an assessment of investment decision times, governance effectiveness, investment capacity, and the ability to lead financing rounds.

The lesson is not that Europe should abandon public equity investment. It is that existing institutions must demonstrate that they can operate effectively before the EU creates another mechanism to perform similar functions.

Any European AI investment framework should incorporate four institutional lessons:

• Time-bound investment decisions: Companies competing in fast-moving markets cannot wait indefinitely for administrative approval.
• Clear decision-making authority: Investment responsibilities must be defined, with accountability for delays and outcomes.
• Professional investment capacity: Public investment institutions need experienced personnel capable of assessing technology, commercial prospects, financial risk, and international competition.
• Authority to lead financing rounds: Where commercially and strategically justified, European institutions must be able to anchor investments rather than depend entirely on another investor taking the lead.

The objective should be to improve Europe's existing investment machinery, not simply create another institution.

The governance test

The challenge extends beyond technology funding. Any new European investment mechanism will depend on the Union's ability to administer large financial commitments, maintain effective oversight, and demonstrate that public resources produce their intended results. This is a broader institutional question, not one confined to AI.

The EU must demonstrate that it can maintain effective financial controls, transparent reporting, and independent oversight when administering large and politically sensitive programs. That capacity will be essential to securing public confidence in any new European AI investment vehicle.

The EU's financial support for Ukraine provides a relevant, although politically distinct, example of the governance challenges involved in administering large programs.

This examination is not a judgment on the war or on the strategic necessity or moral justification of support. Reasonable people may reach different conclusions about the fiscal wisdom, effectiveness, and wider consequences of that support. The question here is narrower: whether the institutions responsible for administering public resources can demonstrate transparency, effective controls, accountability, and measurable outcomes.

The scale and complexity of support to Ukraine have involved grants, loans, guarantees, procurement, and financial instruments operating across multiple institutions and national authorities. That complexity makes clear reporting and independent oversight particularly important.

There have also been specific cases requiring intervention. OLAF identified serious procurement, competition, transparency, and financial-management breaches in an EU-funded generator procurement project administered by Poland's Government Agency for Strategic Reserves. It recommended recovery of more than €91 million and reported that a further €22 million had been safeguarded from improper spending (OLAF, 2025).

This case should not be treated as proof that all Ukraine funding was mismanaged, nor should it be conflated with broader EU budget audit error rates. It does, however, illustrate why emergency spending and politically sensitive financial commitments require strong procurement controls, independent investigation, and effective recovery mechanisms.

The relevant lesson for AI investment is straightforward. Citizens should not be expected to accept greater financial exposure simply because an initiative has strategic importance. Institutions must establish clear objectives, demonstrate their ability to control expenditure, report results, identify failures, and correct them.

Accountability, transparency, and measurable results are necessary conditions for institutional legitimacy. They do not replace political debate about whether a program should exist or what objectives it should pursue.

What Europe can learn from its existing institutions

Europe already has several mechanisms that could contribute to a more effective AI investment and industrial strategy.

The European Innovation Council Fund provides a model for direct equity investment in deep-technology companies. InvestEU and the European Investment Fund offer experience in mobilizing private capital through financial intermediaries, venture capital, growth equity, and other instruments. The European Tech Champions Initiative has sought to address the shortage of large-scale venture financing for European technology companies (European Investment Fund, 2023; European Investment Fund, n.d.-a, n.d.-b).

The European Chips Act provides a particularly relevant industrial comparison. It combines research support, public finance, state aid, and investment incentives to strengthen semiconductor manufacturing and supply-chain resilience.

The Commission has reported substantial investment commitments under the initiative. However, the European Court of Auditors concluded in 2025 that the Chips Act was very unlikely to achieve the EU's target of 20% of global semiconductor production value by 2030. The auditors identified limitations in the Commission's resources and mandate, alongside the importance of Member State and private-sector investment (European Court of Auditors, 2025).

The lesson is important. Europe can establish industrial programs, mobilize substantial investment, and create new infrastructure. But those activities do not automatically produce the scale or market position originally intended.

These institutions should not be treated as interchangeable, and their funding figures should not be added together as though they represent one common investment pool. They address different financing needs and operate under different mandates.

The more realistic approach is to improve coordination between them.

Europe needs a system in which research funding can connect to equity investment, equity investment can connect to growth financing, infrastructure can support commercial deployment, and procurement can create demand for European products.

That requires clear responsibilities, common performance measures, and mechanisms that allow companies to move between funding and investment stages without repeatedly encountering fragmented administrative processes.

A European model, not Silicon Valley

The answer is not to create a European clone of Silicon Valley.

Europe does not need to reproduce American venture capital culture, concentrate technological development in one dominant hub, or imitate the extreme concentration of capital and market power associated with the US technology sector.

It needs an industrial system suited to its own economic structure, institutional arrangements, and comparative advantages.

That means a distributed but coordinated AI ecosystem.

Strategically significant infrastructure should be developed where scale, energy availability, research capability, industrial demand, and security requirements justify it. Access to that infrastructure, however, should extend across Member States.

A distributed model does not mean distributing every investment equally. Nor does it mean creating an AI data center in every country regardless of commercial or technical viability.

It means coordinating investment in strategically important infrastructure while distributing access, research participation, commercial opportunities, and economic benefits across the Union.

Europe should build on areas where it already has meaningful industrial capabilities: advanced manufacturing, automotive and aerospace engineering, pharmaceuticals, telecommunications, energy, scientific research, and public services.

The objective should be to integrate AI into these industries, improve their productivity, and create commercially successful European technology companies.

A coordinated system would require shared infrastructure standards, interoperable computing and data environments, cross-border investment mechanisms, common procurement principles, and a more effective route from research to commercialization.

This is a more realistic ambition than attempting to reproduce the entire American or Chinese AI ecosystem.

Protecting European savings and public capital

Mobilizing European savings for AI investment is a legitimate objective. It is not, however, an entitlement for governments or European institutions.

Pension funds, insurers, and other institutional investors operate under fiduciary obligations. Their participation must be commercially justified and consistent with their responsibilities to beneficiaries. Household savings should not become a source of politically directed investment or an indirect means of transferring speculative technology risk to citizens.

The same applies to public funding. Taxpayers should not be asked to absorb open-ended investment risk without a clear explanation of the strategic purpose, financial exposure, and expected public benefit.

Before mobilizing substantially more private or public capital, Europe must demonstrate that its existing institutions can operate effectively.

A credible framework would require:

• Commercial and political separation. Governments should establish strategic priorities, but professional investment managers should select individual investments against published criteria.
• Clear risk allocation. Public guarantees, blended finance, and subordinated investment should be used only where a demonstrable market failure justifies public participation.
• Independent oversight. Investment decisions should undergo professional due diligence, conflict-of-interest controls, and independent review.
• Protection of savers. Pension funds and insurers should participate only through structures consistent with fiduciary duties, diversification requirements, and risk limits.
• Transparent reporting. Capital commitments, actual investments, financial performance, and strategic outcomes should be reported separately.
• Cross-border participation. Allocation should be based on commercial and strategic merit rather than national bargaining or political preferences.
• Measurable performance. The framework should report on commercial growth, infrastructure capacity, private capital mobilization, industrial deployment, productivity, security resilience, and financial returns.

These safeguards are not administrative details to be resolved after capital has been committed. They are prerequisites for establishing the credibility of the investment framework itself.

The next moves

Europe does not necessarily need another institution. It needs to make its existing institutions work together more effectively.

A European AI Capital Compact could provide a common financing and coordination framework, building on the EIC Fund, European Investment Fund, InvestEU, national development banks, and private institutional investors.

Its purpose would be to address the financing gaps that prevent promising European AI companies and infrastructure projects from reaching commercial scale.

Rather than establishing an entirely new fund with overlapping responsibilities, the Compact could coordinate existing mechanisms around a clearly defined investment strategy.

That strategy should focus on several areas:

1. Growth and scale-up financing.

European companies need access to substantial equity financing beyond the early stages of research and development. Existing investment mechanisms should be better equipped to support successive financing rounds, attract international co-investors, and lead strategically important investments where justified.

2. Critical AI infrastructure.

AI Factories and Gigafactories should be evaluated against actual demand, technical requirements, operating costs, energy availability, and commercial sustainability. Their purpose should be to provide reliable and affordable computing capacity, not to become prestige projects measured by construction announcements.

3. Industrial commercialization.

Europe should connect research institutions, startups, established industrial companies, and technology investors. The objective should be to turn research and engineering expertise into products and services that can compete internationally.

4. Strategic procurement.

The EU and Member States should use public procurement to create predictable demand for secure, interoperable AI systems. Hospitals, public administrations, utilities, transport operators, and other public institutions can provide valuable reference customers for European companies.

Procurement should remain competitive, transparent, and technically rigorous. European origin should not become a substitute for quality, security, or commercial value. Instead, procurement requirements should reward interoperability, data protection, security, resilience, and the ability to integrate with European infrastructure.

5. Regulatory implementation.

The EU should make regulatory implementation simple, predictable, proportionate, and consistent across Member States.

The AI Act and other relevant regulations can provide a framework for trustworthy AI. But regulatory requirements must be accompanied by practical guidance, consistent enforcement, and manageable compliance processes (European Parliament and Council of the European Union, 2024).

The objective should be to make Europe a place where companies can develop, test, deploy, and commercialize trustworthy AI without encountering unnecessary administrative fragmentation.

These priorities should be pursued through existing institutions wherever possible. Any new financing mechanism should address a demonstrated gap that current institutions cannot reasonably resolve.

The strategic choice

Europe's strategic choice is not simply whether it can match the United States or China across every layer of the AI technology stack. That outcome remains uncertain, and attempting to achieve parity in every area would not necessarily represent the best use of European resources.

The more immediate question is whether Europe can develop enough capability to protect its market, preserve access to critical technologies, support its industrial base, and retain influence over the systems on which its economy and security increasingly depend.

Without a functioning AI ecosystem, Europe risks remaining primarily a rule-maker, customer, and dependent user of technologies developed and controlled elsewhere.

Regulatory influence has limits when the regulated technology is designed, owned, financed, and operated outside the jurisdiction seeking to influence it. Market size matters, but market access alone does not guarantee control over technological development or the distribution of economic value.

Europe must therefore become more than a large market with a sophisticated regulatory framework. It must demonstrate the ability to finance, build, integrate, and deploy AI capabilities at commercially meaningful scale.

Conclusion

Europe's AI challenge is not a shortage of ambition, research, savings, or industrial expertise. It is the difficulty of bringing those resources together to create a competitive and strategically resilient ecosystem.

Christine Lagarde's warning about European savings financing innovation elsewhere captures one part of the problem. The OECD's investment figures illustrate the scale of the financing imbalance. The European Court of Auditors' assessments of AI investment and the Chips Act expose the gap between institutional ambition and measurable outcomes.

These are not arguments for abandoning European industrial policy or accepting permanent technological dependence. They are arguments for making European institutions more capable of delivering what they promise.

Europe should not begin by asking citizens and institutional investors to commit more capital to another ambitious program. It should begin by demonstrating that existing institutions can make sound investment decisions, coordinate effectively, deliver infrastructure, support commercially successful companies, and account transparently for the resources entrusted to them.

The appropriate response is not another layer of European bureaucracy or an attempt to reproduce Silicon Valley. It is a more disciplined, coordinated, and commercially grounded use of the institutions, capital, research, and industrial capabilities Europe already possesses.

Europe does not need to dominate every part of the AI value chain. It needs to become indispensable in enough of it to protect its interests, retain meaningful strategic influence, and ensure that the economic benefits of AI development are not captured predominantly elsewhere.

The question is no longer whether Europe wants to remain relevant. It is whether it can demonstrate the institutional and industrial capability to do so.

References

European Commission (2025a), AI Continent Action Plan, Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, COM(2025) 165 final, Brussels, April 9, 2025, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52025DC0165.

European Commission (2025b), “EU launches InvestAI initiative to mobilise €200 billion of investment in artificial intelligence”, press release, February 11, 2025, https://digital-strategy.ec.europa.eu/en/news/eu-launches-investai-initiative-mobilise-eu200-billion-investment-artificial-intelligence.

European Commission (n.d.), AI Factories, https://digital-strategy.ec.europa.eu/en/policies/ai-factories (accessed on November 9, 2026).

European Court of Auditors (2024), EU Artificial Intelligence Ambition: Stronger Governance and Increased, More Focused Investment Essential Going Forward, Special Report 08/2024, https://www.eca.europa.eu/en/publications/SR-2024-08.

European Court of Auditors (2025), The EU’s Strategy for Microchips: Reasonable Progress in Its Implementation but the Chips Act Is Very Unlikely to Be Sufficient to Reach the Overly Ambitious Digital Decade Target, Special Report 12/2025, https://www.eca.europa.eu/en/publications/SR-2025-12.

European Innovation Council and SMEs Executive Agency (2025), EIC Impact Report 2025: Key Impacts in Support of Deep Tech Innovation, https://eic.ec.europa.eu/document/download/7b947b36-66cb-4471-a2d0-158d5ae6770f_en?filename=EIC-Impact-Report-2025.pdf.

European Investment Fund (2023), “Launch of New Fund of Funds to Support European Tech Champions”, press release, February 13, 2023, https://www.eif.org/press/all/launch-of-new-fund-of-funds-to-support-european-tech-champions.

European Investment Fund (n.d.-a), ETCI Overview: The European Tech Champions Initiative, https://www.eif.org/flagship-initiatives/european-tech-champions-initiative/overview (accessed on November 9, 2026).

European Investment Fund (n.d.-b), InvestEU overview, https://www.eif.org/flagship-initiatives/investeu/overview (accessed on November 9, 2026).

European Parliament and Council of the European Union (2024), Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act), https://eur-lex.europa.eu/eli/reg/2024/1689/oj. Blue Ocean Labs library record: BOL-R-0003.

Lagarde, C. (2026), “A new age of capital: Growth, sovereignty and AI”, speech at “Hofburg im Dialog – Economy, Europe, Resilience”, Vienna, September 14, 2026, European Central Bank, https://www.ecb.europa.eu/press/key/date/2026/html/ecb.sp260914_2~a3f0efbee4.en.html.

OECD (2026), “Venture capital investments in artificial intelligence through 2025”, OECD Policy Briefs, No. 50, OECD Publishing, Paris, https://doi.org/10.1787/a13752f5-en.

OLAF (2025), “OLAF completes investigation into suspected serious irregularities in EU-funded procurement of power generators for Ukraine”, Press Release No. 5/2025, European Anti-Fraud Office, February 18, 2025, https://anti-fraud.ec.europa.eu/media-corner/news/olaf-completes-investigation-suspected-serious-irregularities-eu-funded-procurement-power-generators-2025-02-18_en.

Shanghai Media Group (2025), [ShanghaiEye: International Communication and Global Reach], https://www.smg.cn/review/202506/0166515.html.

ShanghaiEye (2026), “A very bold ECB Chief Lagarde on Europe's money, AI and China trade ties”, video interview with Christine Lagarde, conducted with China Media Group, YouTube, November 2, 2026, https://www.youtube.com/watch?v=B9_crPW3Hj4. Source transparency note: ShanghaiEye is an international communication brand of SMG International, established by Shanghai Radio and Television (Shanghai Media Group, 2025). According to State Media Monitor, Shanghai Media Group is a wholly state-owned cultural and media group operating under Shanghai municipal oversight (State Media Monitor, 2026). ShanghaiEye is cited here only as the publication record of the interview, not as an independent authority on EU policy.

State Media Monitor (2026), “Shanghai Media Group (SMG)”, August 8, 2026, https://statemediamonitor.com/2026/08/shanghai-media-group-smg/.