by Xinhua writer Shao Haijun
ZAANDAM, THE NETHERLANDS, Oct. 12 (Xinhua) -- Europe's ambition to strengthen its artificial intelligence (AI) capabilities faces challenges, from dependence on foreign technology providers and limited investment in domestic companies to emerging safety risks and regulatory uncertainty.
At the recent World Summit AI 2026 in Zaandam, the Netherlands, participants said Europe needed to expand its computing infrastructure, channel more capital into domestic technology companies and build a competitive ecosystem giving governments and businesses greater control over the AI systems they use.
TECHNOLOGICAL DEPENDENCE LIMITS EUROPE'S CONTROL
A fundamental challenge for Europe is reducing its dependence on foreign technology providers without sacrificing the benefits of international cooperation and open markets.
Mark Papermaster, chief technology officer of U.S. semiconductor company AMD, said on the opening day of the summit on Oct. 7 that AI sovereignty meant having control over computing infrastructure and the freedom to choose among different technologies, combine components and use both open-weight and proprietary models.
For European governments and businesses, dependence on a limited number of suppliers can constrain their choices in developing, deploying and securing AI systems.
Nigel Shadbolt, a member of the British government's Council for Science and Technology, stressed the importance of that choice. "When we talk about sovereignty in Europe, it's about having the ability to make a choice," he said. "You haven't locked yourself into a particular dependency."
Sovereignty, however, does not necessarily mean producing every component domestically.
Papermaster said an open ecosystem could allow European governments and businesses to combine different hardware and software components and models, provided appropriate security measures and safeguards were built into the systems.
LIMITED FINANCING THREATENS AI COMPANIES' GROWTH
Financing remains another major obstacle to turning Europe's strengths in research, talent and computing initiatives into global competitiveness.
"I think one of Europe's biggest weaknesses is on the finance side," said Olivier Debeugny, CEO of OVHai LLM, a French company specializing in resource-efficient AI models and architectures.
Debeugny said Europe had a strong foundation in education and the human capital needed for AI development, but it needs more financial support to turn those advantages into competitive businesses.
He argued that substantial spending on digital services in Europe ultimately benefited U.S. companies. "Every year, a huge amount of money is spent on digital services in Europe. A very large share of that money ultimately goes to companies in the United States," he said.
European Central Bank President Christine Lagarde highlighted a similar pattern in a speech in Vienna on Sept. 14, 2026. "European savers are again financing the transformative technology of the age: artificial intelligence. Again, most of the money is flowing to the United States," Lagarde said, noting that euro area households held about 440 billion euros (492 billion U.S. dollars) in U.S. technology firms.
The challenge is not simply a shortage of capital, but also how existing investment is allocated. Cristina Caffarra, co-founder and chair of the Eurostack Initiative Foundation, said European institutional investors managed trillions of euros, much of which remained invested in traditional assets.
"We need to find ways to channel more of this capital into European technology," Caffarra said.
AI SAFETY RISKS DEMAND STRONGER SAFEGUARDS
The third challenge is ensuring that increasingly capable AI systems remain secure and operate within appropriate boundaries.
Recent incidents involving advanced AI agents have highlighted this concern. In June, an AI agent developed by OpenAI reportedly breached a data portal linked to Australia's universal healthcare system.
"Highly capable systems could, for example, cheat, hack into systems or obtain credentials in ways their developers had not anticipated," Shadbolt said.
Papermaster said addressing these risks required more than improvements to individual AI models. Safeguards also needed to extend across computing infrastructure and the applications built on it.
"Security and safety are fundamental and need to be architected into every facet of AI applications," he said.
As AI systems become more capable of carrying out tasks autonomously, ensuring their security will become increasingly important to maintaining public trust and enabling wider adoption.
REGULATORY UNCERTAINTY MAY WEAKEN EUROPE'S COMPETITIVENESS
Europe also must maintain its influence over global AI governance while creating conditions for domestic companies to innovate and compete.
Sasha Rubel, head of generative AI policy for Europe, the Middle East and Africa at Amazon Web Services (AWS), highlighted the compliance burden facing European businesses and the need to make regulatory requirements clearer and easier to implement.
"The fundamental question is: how do we simplify the rules?" Rubel said. "This is not about whether we should regulate. It is about making the rules clear enough that they can actually be applied and that innovators understand how to comply."
Her remarks reflect industry concerns that regulatory complexity and uncertainty can divert resources from innovation and make it harder for companies to bring new technologies to market.
Daniel Friedlaender, senior vice president and head of office at the Computer & Communications Industry Association Europe, pointed to a related gap between technological innovation and AI adoption.
Europe has considerable innovation in research laboratories and businesses, he said, but uptake has been slower. Regulatory predictability, clarity over compliance and access to capital were among the factors affecting companies' ability to scale up and expand their markets.
Ultimately, AI sovereignty will depend not only on the technologies Europe develops and the rules it establishes, but also on whether it can finance, secure and scale those technologies while retaining the freedom to choose among global suppliers. ■