The divergence effect of the AI economy: Why is Europe more worried about being "excluded from the supply chain"?

The Divergence Effect of the AI Economy: Why Europe Is More Worried About Being “Left Out of the Supply Chain”

AI is pushing the global economy toward a new tiered structure: a small number of countries and companies that control models, chips, computing power, and distribution channels may earn excess profits, while more economies face job restructuring, erosion of the tax base, and rising pressure on social welfare systems. In the face of this transformation, what Europe should worry about most is not simply that white-collar jobs will be automated away, but whether Europe will remain in the upper-middle of the value chain in the next round of industrial restructuring, or slide further toward the margins of technology and profit distribution.

A key judgment in the discussion cited by The Guardian is that the pressures brought by AI in the future will not be evenly distributed. Silicon Valley in the United States is experiencing extreme concentration of capital, talent, and expectations, while South Korea, Japan, and Taiwan are sharing in the gains from the AI investment boom thanks to their key positions in the semiconductor supply chain. Europe’s situation is more complicated. Dutch company ASML remains an indispensable key node in global advanced chip manufacturing, but cases like this are too rare to support Europe’s AI industrial narrative on their own.

This means that Europe’s primary problem in the face of the AI wave is not “whether people will know how to use AI,” but “what position Europe can occupy in the AI industrial chain.” If a region can only consume AI services, while lacking control over chip equipment, cloud infrastructure, foundation models, enterprise software, and industry application entry points, then it will receive only a small share of the efficiency gains, while bearing most of the costs of job displacement, income pressure, and fiscal redistribution.

From the perspective of EU competitiveness, this kind of divergence is especially dangerous. Europe has long emphasized open markets, rule-based governance, and technological neutrality, but the distribution of returns in the AI era is more like a combination of the platform economy and hard-tech manufacturing. Whoever owns training computing power, whoever controls advanced-process equipment, and whoever controls data centers and enterprise application gateways can more easily turn AI from a “tool” into a “machine for industrial profits.” If Europe cannot build stronger domestic capabilities in these links, it may continue to rely on external technology supply in the next round of digital industrial upgrading.

This also explains why the focus of recent EU policy has gradually shifted from digital regulation alone to building industrial capacity. The AI Act represents a rule framework, not an industrial outcome; it can improve transparency and accountability, but it cannot automatically produce Europe’s own AI champion companies. What will truly determine long-term competitiveness are computing infrastructure, the chip ecosystem, data availability, the depth of risk capital, the speed of enterprise adoption, and whether the cross-border market is large enough to support technology commercialization.The second European proposition of the AI economy is fiscal resilience. The original text discusses an easily overlooked fact: if a country cannot obtain enough tax revenue from the AI boom, and has no excess profits to redistribute, then it will find it harder to buffer the shock of automation. For Europe, this is not a theoretical issue. European welfare states depend more on a stable tax base and high employment rates, and if AI mainly replaces mid-level white-collar jobs, what will be affected is not only individual workers, but also consumption, social security contributions, and local economic vitality.

In other words, AI will not simply turn “high-paying jobs” into “low-paying jobs”; more likely, it will directly pull some occupations out of the employment system. If European companies are the first to deploy AI widely in finance, law, software development, customer support, and process management, while domestic new industries do not expand in tandem, then productivity may rise in the short term, but in the long term there may be structural contradictions such as slower wage growth, insufficient incremental tax revenue, and rising pressure on social spending.

This is also why Europe cannot measure AI’s impact solely by labor substitution rates. What really matters is whether AI will bring a new investment cycle. If AI drives linked investment in Europe’s data centers, semiconductor packaging, industrial software, automation equipment, and energy infrastructure, then it may become part of Europe’s reindustrialization. If AI is mainly reflected in external platforms exporting efficiency tools to European businesses, then what Europe gets will be more cost optimization than industrial upgrading.

Energy and infrastructure play a decisive role here. The commercialization of AI depends on computing power, and computing power depends on electricity, cooling, land, networks, and capital density. Europe’s structural constraints in energy prices, approval speed, and grid capacity may directly affect its ability to attract AI infrastructure. For a continent that is trying to advance green transition, industrial reshoring, and digital sovereignty at the same time, this constraint is not a marginal issue, but a competitiveness issue.

The original text also points to another trend with greater global political economy significance: AI will not only create inequality within developed countries; it may also widen the gap between countries. For Europe, this means two risks exist simultaneously. First, the digital capability gap between stronger and weaker countries within Europe may widen; second, industrial differentiation between Europe and the United States, China, South Korea, and Taiwan will deepen further. The former affects EU cohesion, while the latter affects the EU’s negotiating position in the global technological order.

If the winners in the AI era are concentrated in a handful of technology hubs, then Europe must answer a more practical question: what does strategic autonomy really mean? It is not just about reducing dependence on a single supplier, nor is it just about setting stricter regulatory rules; it is about ensuring that Europe has sustainable sources of profit, scalable innovation capacity, and a defensible position in supply chains in the next generation of production systems.From this perspective, ASML-like success is certainly important, but it is not enough to prove that Europe is ready for the AI era. What Europe needs more is a complete industrial ecosystem: from advanced manufacturing equipment and power semiconductors to cloud infrastructure, from industry-specific AI applications for manufacturing, healthcare, automotive, finance, and public services to capital markets and a single-market mechanism that can support startups to scale. Without these links, Europe is likely to become an important maker of AI rules, but not an important allocator of AI value.

This is precisely the part of the AI economy that Europe should be most wary of. On the surface, it is a technological revolution; at a deeper level, it is a reordering of industrial organization, labor market structures, and the state’s fiscal capacity. For Europe, the issue is not whether to embrace AI, but whether it can turn AI from an external shock into an opportunity for internal upgrading. If it cannot, Europe will face not the decline of a single industry, but a long-term situation in which competitiveness, the tax base, and the social contract are all under pressure.

SEO Description From the perspectives of Europe’s competitiveness, EU industrial policy, and AI supply chains, this examines how the AI economy is reshaping global wealth distribution and assesses Europe’s long-term challenges in chips, computing power, regulation, and fiscal resilience.

Source URL https://www.theguardian.com/business/2026/jun/02/will-the-ai-economy-create-a-permanent-underclass