From a strong knowledge base to a digital lead

Digital knowledge will only give the Netherlands and Europe an advantage if it leads to scalable alternatives and organisations are genuinely able to choose them. France’s Cloudwatt shows that a domestic cloud offering does not automatically become a sustainable alternative. Despite support from major industrial partners and €75 million in state investment, the initiative failed to achieve sufficient commercial scale and the platform closed in 2020. Technological availability and national origin are not enough: realistic market positioning, a sufficient customer base and investment capacity are also needed to keep an alternative viable over the long term. Research findings may be commercialised under foreign control, while dependence on foreign chips, cloud capacity and software limits Europe’s own room for manoeuvre.

One example is Finland’s Silo AI, which developed language models with the University of Turku using European research funding. In 2024, the company was acquired by US-based AMD. As a result, the commercial translation of this scientific knowledge into applications came under US control. European knowledge development therefore does not automatically result in European control over its further commercialisation. International cooperation remains important for access to knowledge and markets, but it does not guarantee that Europe can chart its own course. What is needed to turn knowledge into viable alternatives?

Assess critical links as an interconnected whole

Digital autonomy is about the ability to continue using, adapt or replace critical systems. Control over systems, algorithms and data is central to this. A European label alone provides no guarantee. What matters is whether an organisation can, in practice, adapt or replace a system or fall back on another solution.

Critical links must therefore be assessed as an interconnected whole: what knowledge and production capacity are available, which alternative suppliers exist, can systems interoperate, and who has influence over standards? Mapping dependencies is the starting point, not the end result. The next step is to determine where intervention will genuinely deliver greater control. This requires clear agreements on measures, powers and responsibilities. Because critical links can shift, they must be reassessed regularly. This calls not only for insight at national or sector level, but also for sufficient expertise within government and business. Organisations need to be able to assess dependencies, manage suppliers and make effective use of alternatives. Where expertise is concentrated with a single supplier or has largely been outsourced, rebuilding that expertise in-house requires explicit attention.

Turn technological strength into market strength

Good technology does not reach the market by itself. It also requires entrepreneurs, financing, production capacity and first customers. Ideally, research, test facilities and market introduction should not be separate tracks, but parts of a single coherent route from research to application and scaling. Success cannot be measured solely by patents or research results. Access to a sufficient customer base and opportunities for companies to grow also determine whether technology secures a lasting place in the market. Digital sovereignty is therefore not only about limiting risks, but also about the capacity to scale innovations, develop new markets and retain economic value in the Netherlands and Europe.

Initiatives such as EuroStack illustrate the search for stronger European positions across the digital value chain, from chips and digital infrastructure to software and services. Government, businesses and research organisations have complementary roles: government can act as a first customer to open up markets, businesses can scale solutions, and research organisations can test them independently. However, expanding the European offering must not leave organisations dependent once again on just a few major suppliers. European scale does not automatically create European freedom of choice. Digital sovereignty is therefore also a transition challenge. Technological alternatives can only be implemented and scaled structurally if the surrounding system also changes, from financing and procurement to standards, regulation, market incentives and user behaviour.

Build alternatives and switching capacity in parallel

Organisations may formally have freedom of choice, while switching is impractical because of complex data migration, intertwined work processes, the time and effort required to deploy and train staff, or a lack of in-house knowledge and trust in the technology. Reducing dependencies is therefore not a one-off switch: alternatives and switching options must be developed and tested step by step. Switching too early can threaten continuity, while waiting until an alternative is equivalent in every respect can strengthen existing dependencies. This requires an explicit assessment of whether temporary additional costs or limitations are acceptable. New solutions need customers in order to grow. At the same time, customers will only choose a solution if it offers them enough. By procuring jointly, European buyers can help break this cycle. For a sovereign government cloud, this means that cooperation between governmental data centres and clear criteria for deciding when switching is a good solution must also form part of the approach.

Invest in tomorrow’s technology by building on existing strengths

Addressing existing vulnerabilities alone is not enough to secure a strong position in the future. Photonic, quantum and neuromorphic computing, new chip architectures, energy-efficient data processing, open software and secure data infrastructures also deserve attention. The focus should not be solely on individual technologies, but on their position within the wider digital stack and value chain. A strong position in one layer of the stack can increase control and innovative capacity in other layers.

The Netherlands can build selectively on its combination of scientific expertise, high-tech manufacturing and application knowledge. Investment in shared research and production facilities and in talent can reinforce these connections. The Netherlands does not need to develop everything itself. The key question is where its combined existing strengths can make a real difference. This will keep investment in future technologies focused, rather than spreading it too thin across too many areas.

Dependency is also a geopolitical choice

Which dependencies are acceptable is not determined solely by technological and economic considerations. Export controls, conflicts and changing trade rules can rapidly affect access to technology. It is not only the supplier’s country of origin that matters, but also the terms under which cooperation takes place. Do the parties involved have access to one another’s knowledge and markets? What contractual and political conditions apply? Connecting technological, economic and geopolitical information provides a clearer picture of the available choices. This helps determine when international cooperation is appropriate, when multiple suppliers are needed and when developing a digital technology in-house is a better option.

Not every application or system needs to meet the same sovereignty requirements. How sensitive is the data, and what happens if the system fails? These questions help organisations differentiate between cases. However, the acceptable level of risk cannot be determined on technical grounds alone. It also requires clarity about who decides how much dependency is acceptable and who bears the consequences if things go wrong. Linking such assessments to concrete fallback options clarifies which risks an organisation is taking or can reduce, and who is responsible for what.

From dependency to room for manoeuvre

Digital sovereignty is not an end state that can be achieved through a single investment or decision. It remains a balancing act between open collaboration, continuity and control. Digital knowledge only becomes a digital advantage when scalable alternatives are combined with organisations’ ability to use them in practice. Targeted investment is needed, while real-world testing shows whether freedom of choice is genuinely increasing. Ultimately, the measure of success is not whether every dependency disappears, but whether governments and businesses can genuinely make different choices, enabling the Netherlands and Europe to continue charting their own course.