Europe should follow China’s AI diffusion model

Jorge Costa Oliveira

The global debate over artificial intelligence has been dominated by an arms-race mentality focused on the technological frontier. The United States and Silicon Valley measure success through ever-larger language models, computing power and the pursuit of artificial general intelligence.

China, however, is quietly developing a different strategy. Rather than competing exclusively to build the world’s most advanced generative model, Beijing is treating AI as a basic input of production, concentrating on its mass adoption across industry, logistics and distribution.

For Europe, caught between American innovation and Asian scale – and frequently burdened by its own regulatory architecture – this approach offers a pragmatic path towards economic relevance in the AI era.

China’s strategy, reinforced by its 15th Five-Year Plan and “AI+” initiatives, aims to move AI out of laboratories and directly onto factory floors and into services. Restrictions on access to advanced chips have also encouraged Chinese companies to develop smaller, highly specialized models and extract greater performance from less powerful hardware.

The real Chinese bet is therefore not simply on creating smarter AI, but on turning AI into a widespread industrial productivity tool.

Europe should pay attention.

The EU suffers from what might be called “frontier anxiety”: frustration that it has failed to produce an equivalent of OpenAI, Google or Microsoft. Yet competing directly with the enormous American investment in computing infrastructure, or China’s concentrated state support, risks becoming an unwinnable financial battle.

More importantly, it misses where much of AI’s economic value will ultimately be created.

Europe’s greatest strength remains its highly specialized small and medium-sized enterprises and precision manufacturing industries. Helping them deploy predictive maintenance, supply-chain optimization, computer vision and AI-assisted robotics could dramatically improve productivity without requiring them to invent the underlying technology.

AI could also help Europe confront its demographic problem. An ageing population and shortages of skilled labor threaten economic growth and the sustainability of social systems. Automation can multiply the capabilities of existing workers rather than simply replace them.

The same logic applies to Europe’s green transition. AI can improve smart electricity grids, optimize freight routes, reduce energy consumption in buildings and factories and support precision agriculture.

But achieving this requires regulatory adjustment.

The EU’s AI Act provides important safeguards for fundamental rights, but its precautionary approach also creates substantial compliance costs. Europe now needs to complement regulation with active facilitation.

Sector-specific regulatory sandboxes should allow manufacturers, ports, railways and logistics companies to test AI applications under supervision without facing disproportionate legal uncertainty. Shared European industrial data infrastructure should be accelerated, while governments should use public procurement to create demand for locally developed and adapted AI systems.

Above all, compliance must become manageable for SMEs. Pre-approved AI tools and fast-track procedures could allow smaller companies to adopt low-risk applications without needing multinational-sized legal departments.

Europe does not need to copy China’s political system to learn from its industrial strategy. It can preserve democratic safeguards while adopting the principle that matters: diffusion.

There is a fundamental difference between creating AI’s “brain” and applying its “muscles” throughout the economy.

Europe’s success should not be measured by whether it creates the world’s smartest model, but by whether it builds the world’s smartest factories, hospitals, transport networks and energy systems.

European technological sovereignty requires a change of focus: less frontier, more diffusion.

linkedin.com/in/jorgecostaoliveira