Dutch cabinet adopts AI vision with three pillars and a 2027 infrastructure plan

The Dutch Council of Ministers agreed the Cabinet Vision on Artificial Intelligence, *Choosing AI for Everyone*, on 6 November 2026, following a proposal from State Secretary for Digital Economy and Sovereignty Aerdts. The vision sets three implementation pillars: responsible AI development and use; access to advanced AI technology and security; and steering the AI transition in line with public values.

For compliance teams, this is a policy direction rather than a new rule or implementation act. The cabinet says it will help companies and public bodies apply the European AI Regulation, including by allowing supervised experimentation with new AI applications. The announcement gives no new private-sector compliance deadline. Instead, it points to future infrastructure policy, ministerial implementation and possible institutional changes that could affect how firms develop, procure and govern AI in the Netherlands.

Three pillars for implementation

The first pillar focuses on making it easier for Dutch AI companies to grow and for businesses and government bodies to adopt AI responsibly. The cabinet identifies talent, capital, data, computing capacity and clear, workable legislation as necessary conditions.

It will develop a broad approach to data centres and publish an elaboration of the AI infrastructure and computing capacity needed by the Netherlands in early 2027. It also plans to remove barriers that prevent small and medium-sized enterprises from using AI.

Within government, successful AI applications are to be made more widely available and public bodies are expected to purchase more jointly. The stated aim is to prevent each organisation from having to develop the same solution independently.

The second pillar concerns access to advanced AI technology and security. The cabinet says AI can make cyberattacks more sophisticated and can be misused for disinformation and influence operations. It also wants to reduce dependence on a small number of foreign companies for critical technology.

The government will determine which technology and infrastructure the Netherlands and Europe should retain domestically, and where cooperation with reliable international partners is sufficient. It is also examining the possible establishment of a Dutch AI safety institute to identify and address new risks.

The third pillar places AI within the Netherlands’ public-value framework. The cabinet says the European AI Regulation establishes requirements for safe and responsible use and prohibits AI involving unacceptable risks. It intends to help companies and public authorities apply those rules in practice, including through supervised experimentation.

The programme also includes improving public understanding of AI and helping people prepare for its effects on work. The cabinet is working with the media sector on a Media Pact covering reliable information and the use of journalistic content by AI.

Sovereignty without full independence

The cabinet’s stated objective is not complete technological self-sufficiency. It wants enough domestic knowledge and capacity to retain access to important technology and continue making its own choices. In the cabinet’s words, “the goal is not full independence”, translated from the Dutch announcement.

That distinction matters for firms relying on foreign foundation-model providers, cloud infrastructure or other external AI services. The policy points towards a future assessment of which dependencies are acceptable and which capabilities should be available in the Netherlands or Europe.

The implementation challenge is already visible in the regulatory architecture. In its 12 June 2026 implementation assessment, the Dutch Authority for the Financial Markets found the AI Regulation Implementation Act feasible but sought changes concerning supervisory roles, capacity, data sharing and publication regimes.

Ministries will turn the vision into sector work

The vision is presented as the joint course of the whole cabinet. Aerdts will coordinate the approach, while individual ministries will translate it into work covering areas including healthcare, education, the economy and national security.

The cabinet also plans public participation in 2027 and 2028 on the opportunities and limits of AI. That process sits alongside, rather than replaces, the legal obligations created by the European AI Regulation.

The next named implementation event is the cabinet’s elaboration of Dutch AI infrastructure and computing capacity in early 2027.