According to the Financial Times, journalist Soumaya Keynes noted that while Europe lags behind the US and China in developing advanced AI models, it is benefiting from infrastructure investment, equipment manufacturing, and AI applications.
Out of the world’s top 64 large language models, only 1 comes from Europe. Some proposals suggest the EU needs to invest about €800 billion over 3 years to develop advanced AI models capable of competing.
In September 2026, machinery production in the Netherlands increased by 43% year-on-year, largely driven by ASML.The EU’s exports of electrical equipment, cooling systems, and data center operations also surged.
According to Goldman Sachs, the EU accounts for about 10% of global new data center capacity in 2026. Business leaders report that AI helps increase capital expenditure by an average of 10%.
However, Goldman Sachs estimates that new AI infrastructure investment contributes about 0.1 percentage points to European economic growth.
Europe has also increased software license imports from the US, impacting the current account surplus and raising questions about the economic benefit distribution from AI.
A European Investment Bank survey shows that the proportion of businesses using AI is nearly equivalent to the US.However, US businesses apply AI more broadly to drive revenue growth, while European businesses prioritize cost reduction and efficiency improvement.
Research shows that when knowing domestic competitors are increasing AI investments, European businesses tend to upgrade their own investment plans.
📌 Conclusion: Europe does not lead in AI models but is deeply engaged in the value chain through semiconductor equipment, data center infrastructure, and enterprise applications. Although new AI infrastructure investment contributes about 0.1 percentage points to growth, the corporate AI adoption rate is now nearly equivalent to that of the US. The lesson for Vietnam is that it is not strictly necessary to lead in foundational AI model development to reap economic benefits. More important is the capability to participate in the supply chain and translate AI applications into actual productivity.

