ISF Voices 2026: From Regulatory Superpower to Innovation Engine: A Strategic Pivot for the EU’s AI Sovereignty
The latest edition of ISF Voices.
In this special edition of SCSP’s newsletter, we continue our ISF Voices series, showcasing writing by alumni and fellows in SCSP’s International Strategy Forum (ISF) program. Each piece reflects the unique vantage points of emerging leaders from around the world working to shape the future of geopolitics, technology, and democracy. Today’s piece is by David Timis, a 2025 ISF Fellow and a Global Communications & Public Affairs Manager at Generation, one of the largest skilling and employment nonprofits in the world. In this article, David discusses how the European Union can undergo a strategic pivot to develop a new model for European AI sovereignty to secure the Union’s relevance in the global technology race.
The views expressed in this article are those of the author and do not necessarily reflect those of the Special Competitive Studies Project (SCSP), the International Strategy Forum (ISF), or any affiliated persons and institutions.
I. Introduction
One year after the release of the landmark report on European competitiveness by Mario Draghi, the European Union (EU) finds itself at a historic crossroads. This moment is defined by a profound geopolitical awakening, a realization that the continent’s strategic dependencies have moved from theoretical vulnerabilities to near-existential threats. For decades, Europe operated under the assumption that global trade and technological interdependency were shields against conflict; instead, they have become conduits for coercion. The hard truth facing Brussels today is twofold: first, Europe depends on rival powers for the very components of its future survival, and second, those powers are increasingly willing to use their technological strangleholds to bend Europe to their wills.
While the Draghi Report served as a necessary wake-up call for the continent’s existential productivity gap, the subsequent response has been characterized more by bureaucratic deliberation than by the radical structural shifts that are needed to not just be competitive, but stay relevant. As the global artificial intelligence (AI) race intensifies, Europe appears to have woken from its innovation coma only to find itself in a regulatory straitjacket. This internal paralysis is particularly stark when contrasted with the aggressive industrial policies of the United States and China. While Brussels debated the ethical nuances of various AI risk tiers, the U.S. private sector and the U.S. Government accelerated the deployment of frontier models, and Beijing integrated AI into the very fabric of its state-owned enterprises.
As the Financial Times (FT) has noted, this awakening has stripped away the illusion of “strategic autonomy” as a purely regulatory exercise. True autonomy cannot be drafted in a directive; it must be built in a foundry. When those who control the high-end semiconductor supply chain or the primary energy inputs for the next industrial revolution decide to prioritize their own national interests, a Europe that only knows how to regulate finds itself toothless. The “Brussels effect,” the long-held belief that the EU’s regulatory superpower could dictate global standards, is hitting a wall of geopolitical reality. Regulation without innovation is not leadership; it is a retreat into a smaller, less relevant corner of the global economy.
The sobering reality is that while the United States produced 40 large foundation models last year and China produced 15, the EU managed a paltry 3. This disparity is not merely a statistical anomaly; it is the result of a systemic design flaw in Brussels, a structural bias that prioritizes the creation of rules and bureaucratic friction over the fostering of investment that scales breakthroughs from the lab to the market. Today, the United States leads in frontier research and capital markets, while China dominates in physical infrastructure and rapid technology diffusion. In contrast, Europe is increasingly regulating itself into irrelevance.
Christophe Fouquet, the Chief Executive Officer (CEO) of ASML, Europe’s most valuable public company, recently issued a stark warning: while the world debates the victor of the AI race, Europe is definitively losing it. Fouquet noted that only 2% of ASML’s total business stays within Europe, a figure that has remained stagnant for a decade. This is not merely a manufacturing gap; it is a consumption crisis. Because the hyperscalers and designers developing advanced AI chips are located in the United States, there is no domestic ecosystem to absorb the very lithography technology Europe produces. To remain a global player, the EU must move beyond regulatory debates and focus on the major task of attracting the entire ecosystem of AI models, chips, and talent back to European soil.
II. The Pivot to Vertical AI
The conventional wisdom in Brussels often suggests that Europe must build a “Euro-ChatGPT” to achieve digital sovereignty. This is a strategic fallacy. Chasing the United States in the development of general-purpose frontier models is a race Europe has already lost, primarily due to a lack of high-end Graphics Processing Unit (GPU) supply, designed by the likes of NVIDIA, Broadcom, and AMD, and the sheer scale of American private capital. As Mario Mariniello of Bruegel argues, Europe’s true path to sovereignty lies not in replicating the American model, but in a “below the frontier” strategy that prioritizes the adoption and integration of AI into the continent’s existing industrial moats. Instead of betting taxpayer billions on building AI factories that struggle to compete with U.S. hyperscalers, the EU should focus on vertical AI: fine-tuning specialized models that leverage Europe’s high-quality industrial datasets in manufacturing, healthcare, and renewable energy.
By shifting from a “regulation-first” to an “application-first” mindset, Europe can solve the productivity puzzle that even the tech leaders in the United States have yet to crack, transforming raw compute power into tangible industrial gains. In this context, Europe’s path to relevance lies in applying artificial intelligence to its deep domain expertise and high-quality industrial datasets. The continent remains a global powerhouse in manufacturing, healthcare, and renewable energy; these sectors possess vast repositories of specialized data that consumer-facing AI giants cannot easily access. By focusing on specialized AI applications—tools designed to optimize a power grid, revolutionize drug discovery, or automate complex industrial processing—Europe can build a strategic moat. This “application moat” serves as the necessary digital bookend to Europe’s existing physical moat, such as ASML’s monopoly on extreme ultraviolet (EUV) lithography. Together, these upstream technological chokepoints and downstream domain expertise create a unique defensive perimeter that raw compute power alone cannot breach.
This shift to vertical AI is more than a commercial strategy; it is a mechanism for exerting geopolitical influence. As highlighted in the recent paper Europe’s Trump Cards by Dezernat Zukunft, Europe holds significant leverage through asymmetric dependencies that remain largely untapped in the digital age. For example, European suppliers such as Urenco (the Netherlands and the U.K.) and Orano (France) provide nearly 80% of U.S. uranium imports, and Siemens dominates the medium-sized gas turbines required to power the U.S. data centers running frontier models. By integrating AI into these specialized industrial value chains, Europe can transition from being a mere consumer of foreign technology to an indispensable provider of high-value industrial intelligence. This new game moves the goalposts from a race for general intelligence to a race for industrial integration.
III. From Scarcity to Growth Infrastructure
The shift toward industrial integration described above cannot happen in a vacuum; it requires a physical foundation that Europe currently lacks. The defining bottleneck of the AI era is no longer just code or datasets, it is access to abundant, reliable, and cheap electricity. For decades, European energy policy has been defined by “negawatts,” a pursuit of efficiency and waste reduction. While noble, this mindset is fundamentally ill-suited for a century where every industrial process is being plugged into a power-hungry grid.
Strategic autonomy in 2026 is synonymous with computational and energy sovereignty. However, the scale of European ambition remains insufficient. While the EU’s recently unveiled “Apply AI Strategy“ and its commitment to “AI Gigafactories“ are necessary steps, their €2 billion budget is a rounding error compared to the U.S. landscape. Across the Atlantic, the U.S. Government is supporting private-sector projects like “Stargate,” a $500 billion infrastructure play that dwarfs European efforts. If Europe continues to fund its future with “symbolic” budgets, it will remain a tenant in an American-built digital world.
To avoid this fate, Europe must shift its energy story from efficiency to growth. Global data center demand is projected to more than double by 2030, and the United States is already widening the gap in hyperscale capacity. Yet, while the race for massive training clusters is a challenge, the race for AI inference, the deployment and real-time use of AI in factories, hospitals, and grids, is just beginning. Inference requires connectivity-focused facilities and massive amounts of fiber, areas where Europe still maintains a competitive edge.
To seize this moment, the EU must resolve the internal paralysis of its physical infrastructure. Grid congestion is currently “freezing” innovation hubs in Germany, the Netherlands, and the United Kingdom, preventing new data centers from coming online. If the “Trump Cards” of European industry are to be played, they must first be powered. Moving from a scarcity mindset to an abundance mindset is the only way to ensure that the “new game” of industrial integration is played on European soil.
IV. Radical Structural Reform
Finally, Europe cannot innovate by decree. We must shift from a “regulation-first” mindset to one that is “innovation-first.” This requires three radical structural reforms.
First, the EU Must Unshackle Capital. The fragmented nature of European capital markets starves our firms of the private investment needed to scale. Completing the Capital Markets Union (CMU) is no longer a bureaucratic choice; it is a survival necessity. In 2025, firms in the United States attracted approximately $194 billion in AI-related venture capital, nearly 75% of the global total. In contrast, the EU27 managed only $15.8 billion. While experts at the IMF and the European Central Bank caution that a unified market alone may not immediately close this ten-fold funding gap, they estimate it could unlock up to €470 billion in additional private capital annually. By harmonizing insolvency laws and mobilizing European pension funds, the CMU would provide the “exit environment” necessary to keep our most promising startups from fleeing to the deeper, more integrated liquidity pools of the United States.
Second, the EU Must Prioritize Action over Reports. The “European AI Act,” while well-intentioned, risks becoming a bureaucratic barrier that forces legislators to admit error only when it is too late. With the most restrictive provisions set to take effect in August 2026, Brussels must ensure that implementation is streamlined to favor adoption and public procurement over friction. The EU needs a “regulatory pause” followed by a strategic pivot that treats technology adoption as a national security priority rather than a compliance exercise. This requires a shift toward aggressive “regulatory sandboxing,” allowing European startups to iterate in real-world environments without the threat of preemptive, innovation-killing fines. Instead of focusing on the theoretical risks of 2030, the Union should be leveraging the AI Act to create the world’s most efficient fast track for industrial AI deployment, ensuring that the rules of the road do not become a dead end for Europe’s most promising founders.
Third, the EU Must Focus on Retaining Talent. The “brain drain” of top European researchers to American big tech companies is a symptom of a broken bridge between academia and industry. While Europe produces nearly 18% of the world’s top-tier AI minds, only 10% choose to build their careers within the Union, driven away by a compensation gap and equity structures that European tax codes continue to treat punitively. We must simplify the path from research labs to the global market, ensuring that European-made breakthroughs remain European-owned. This requires modernizing decaying research infrastructure and reforming archaic intellectual property (IP) rules that often leave university spin-offs trapped in negotiation limbo for years. By the time a European researcher navigates the bureaucratic maze of technology transfer, their U.S. counterpart has already secured the capital and commercial freedom to reach the market. If a researcher in Paris or Berlin finds it easier to commercialize their work in Silicon Valley than in their home city, Europe is effectively exporting the intellectual fuel required for its own industrial renewal.
V. Conclusion
Europe’s third act must be defined by action, not by further elaborated reports or never-ending summits. The continent stands at a point of no return: we are currently being out-spent, out-resourced, and out-scaled by both the United States and China. If Brussels fails to resolve the structural paralysis, the fragmented capital markets, the regulatory friction, and the self-imposed energy scarcity, then Europe will indeed become little more than a museum of 20th-century industrial greatness, regulated by a body that has lost its relevance.
The path forward requires a cold-eyed embrace of reality. We can continue to pursue symbolic ownership of general-purpose models we lack the compute to train, or we can master the “new game” of vertical integration. By leaning into our existing structural moats, EUV lithography, specialized industrial data, and high-value energy value chains, Europe can transition from a vulnerable consumer to a key architect of the global AI economy.
Digital sovereignty will not be won by mimicking the Silicon Valley model, but by perfecting a European one: an ecosystem where domain expertise is the primary moat and industrial intelligence is the primary export. As the Geostrategic Europe Taskforce aptly puts it, Europe must broaden its perspective from “managing partnerships to projecting power.” By unshackling our talent and playing to our structural strengths, EUV lithography, specialized semiconductors, and industrial R&D, we can plug our industry into a grid of abundance and break free from the regulatory straitjacket. We must choose this path of action now, and act with the strategic urgency and singular focus required to secure our sovereign AI future.
If you’ve been following the rapid-fire shifts in the tech landscape, you know that reading about AI is one thing, but seeing it in action is another entirely. That’s why you need to be at SCSP’s 2026 AI Expo.
We are bringing together the brightest minds in national security and private-sector innovation to showcase how AI is actively reshaping our world today.
Why you should register:
Hands-on Tech: Experience the latest in generative models, robotics, and dual-use technologies before they hit the mainstream.
High-Stakes Dialogue: Hear from the leaders who are navigating the AI competition in real-time.
Join the Conversation: Connect with a community of builders, thinkers, and early adopters who are as obsessed with the future as you are.
It’s free to attend!
Exhibitor space is still available! We invite organizations to join our exhibit floor and contribute to the future of innovation. Learn more at https://expo.scsp.ai/exhibit/.