Artificial Intelligence (AI) is redefining technological, economic and security landscapes. The transformative technology introduces new levels of disruption and reorders traditional power balances. At the same time, it offers possibilities for increased productivity, efficiency and automation. The Netherlands and the European Union (EU) face the dual challenge of leveraging AI’s potential while safeguarding national security, democratic resilience and strategic autonomy in an environment mostly dominated by the United States and China.
Building on ten foundational insights from expert conversations – capturing the political, technical and societal dynamics shaping today’s AI landscape (the ‘baseline’) – this report explores three scenarios for how AI could affect Dutch national security over the next five years. These scenarios – so-called ‘Plausible Tomorrows’ – rest on two key uncertainties: (1) the extent to which the Netherlands, the EU and the public can access advanced AI, for example via open-source models and proliferation; and (2) how ubiquitous the technology becomes, from its integration into everyday devices to whether agentic AI[1] fulfils its promises.
Plausible Tomorrow 1 envisions widespread proliferation of open-source AI, empowering both state and non-state actors, but fuelling disinformation, cyberattacks and semi-autonomous terrorism. Plausible Tomorrow 2 highlights the rise of agentic AI and its disruptive integration into all spheres, intensifying geopolitical competition, labour market upheaval and social unrest. Plausible Tomorrow 3 considers a bifurcated world where the EU’s digital dependency deepens and the bloc comes to rely fully on either US or Chinese AI systems, effectively forfeiting its digital sovereignty and resilience irreversibly.
Across all scenarios, AI destabilises the information environment and erodes trust in democratic institutions; creates risks and accelerates vulnerabilities in critical sectors such as defence and health; and imposes significant ecological and energy burdens. At the same time, AI offers opportunities in sectors where the Netherlands holds distinct strengths, or niches, such as AgriTech and semiconductors.
Meeting this challenge requires strategic foresight rooted in, and guiding, a long-term strategic vision for AI; proactive policymaking that sets the conditions for European companies to thrive; and investment at a scale that brings Dutch and European research and development (R&D) spending in line with leading peers, namely at or above the 3 per cent of GDP benchmark – and all while strengthening national security and resilience.
Key recommendations that emerge from our analysis of the three Plausible Tomorrows include:
Control dual-use autonomous systems to mitigate security and ethical risks;
Mandate secure and reliable update practices across AI systems;
Safeguard the information environment against manipulation and misuse;
Prioritise niche domains where the Netherlands already holds a competitive edge;
Protect human-critical, non-negotiable domains against harmful algorithmic outcomes;
Strengthen the competitiveness and indispensability of strategic industries in the Netherlands and EU through domestic and sovereign computational power, resilient deep-tech supply chains and EU-aligned funding.
In short, safeguarding democratic institutions, strengthening Europe’s AI capabilities and shaping global standards for responsible AI are essential for the Netherlands and the EU to secure their place in an AI-driven future.