The European Union Coordinated Plan on Artificial Intelligence is a strategic initiative developed by the European Commission and EU Member States to promote development, deployment and use of AI technologies across the European Union. This chapter discusses how EU Member States are creating the enabling conditions for AI development and uptake in line with the Coordinated Plan. It examines how emerging technologies and shared policy goals have shaped national AI strategies and policies across the European Union. It also looks at governance models used to steer and oversee AI policy, as well as allocation and reporting of public funding for AI initiatives. Finally, the chapter reviews how EU Member States are tapping into the potential of data and fostering critical computing capacity.
Progress in Implementing the European Union Coordinated Plan on Artificial Intelligence (Volume 1)
Abstract
The EU Coordinated Plan on AI is a strategic initiative to promote development, deployment and use of AI technologies across the European Union. It represents a joint commitment between the European Commission and EU Member States to maximise the impact of investments in AI, foster synergies and encourage co-operation across the European Union. The plan outlines a series of concrete actions to facilitate investment decisions, aligning AI policy in the European Union to remove fragmentation. It also aims to contribute to strengthening the global position of the European Union regarding the development and adoption of human centric, sustainable, secure, inclusive and trustworthy AI technologies and applications.
Within this framework, this chapter discusses how EU Member States are creating the enabling conditions for AI development and uptake in line with the Coordinated Plan. It examines how emerging technologies and shared policy goals have shaped national AI strategies and policies across the European Union. It also looks at governance models used to steer and oversee AI policy, as well as allocation and reporting of public funding for AI initiatives. Finally, the chapter reviews how EU Member States are tapping into the potential of data and fostering critical computing capacity.
The EU Coordinated Plan on AI recognised that sharing knowledge, co‑ordinating actions and aligning investments in AI can provide a critical competitive edge. Member States were encouraged to adopt dedicated national AI strategies or incorporate an AI dimension into existing national strategies and share these with the European Commission and with one another. In parallel, the European Commission committed to monitoring developments and mobilising relevant expertise.
Building on the achievements in the first phase of the plan, the 2021 review of the EU Coordinated Plan on AI identified key actions to guide Member States:
Make best use of relevant EU funding possibilities, including the Recovery and Resilience Facility (RRF), to support and reinforce development and uptake of AI technologies at both the national and local levels, based on the national strategies, including by crowding in private investment.
Review and update national AI strategies as necessary to ensure that identified actions and investments are fully realised in practice and inform the European Commission about the progress accordingly.
Develop and promote instruments that allow regular monitoring, co‑ordination, evaluation and exchange of experience and best practice across a broad spectrum of stakeholders.
Reinforce support for and investment in joint actions identified in the EU Coordinated Plan on AI.
Share, develop and implement actions at the national/regional level that proved to be successful in other Member States, for example, successful national initiatives to develop and promote a virtual warehouse of data.
This section discusses the actions by EU Member States to acquire, pool and share policy insights, in line with the EU Coordinated Plan on AI. Table 2.1 summarises key findings from the survey and complementary interviews.
National AI strategies and policies evolve across the European Union, shaped by emerging technologies and shared policy goals
As of December 2024, 24 of the 27 EU Member States had adopted national AI strategies, demonstrating a strong and growing commitment to AI development, governance and adoption across the European Union (Table 2.2). EU Member States exhibit different levels of progress in implementing their national AI strategies. Early adopters have already achieved significant results and advanced substantially. Others remain in the initial stages or have paused or revised strategic approaches due to shifting priorities. Since the revision of the EU Coordinated Plan on AI in 2021, most national AI strategies have been updated or are being reviewed to align more closely with emerging technological trends, regulatory developments and national priorities. In particular, the rise of generative AI has elevated AI on national policy agendas. Recent updates in 2023 and 2024 focus on expanding AI infrastructure, fostering public-private collaboration and addressing the opportunities and risks presented by generative AI.
The EU Coordinated Plan on AI has informed policy in most EU Member States
National AI strategies differ, but the EU Coordinated Plan on AI has influenced most EU Member States in some way. While national AI strategies differ in their priorities, governance models and levels of AI readiness, the EU Coordinated Plan on AI has informed the strategic orientation of AI policymaking and investment in most EU Member States. The vast majority (21) indicated that they considered the plan in some, most or all parts of their strategies. Several EU Member States also mentioned the Digital Decade programme (EC, 2025[1]), including its targets, as a useful reference. Although they widely regarded overarching goals of the EU Coordinated Plan on AI as valuable, national authorities expressed the need for more detailed, practical guidance on how to achieve these objectives. In practice, the content of national AI policies remains largely driven by domestic agendas, with limited evidence of structured cross-border or regional collaboration. Notable exceptions include areas of strategic EU interest that have been important enablers of cross-border co‑operation. These include cloud computing, high-performance computing (HPC) and health-related initiatives like the 1+Million Genomes and Genomic Data Infrastructure projects.
Many EU Member States have updated, or are updating, their national AI strategies
The evolution of national AI strategies across the European Union reflects rapid advancements in AI technology and lessons learnt from prior implementation. Many countries that introduced their strategies before 2020 have updated them to ensure alignment with AI advancements, regulatory changes and national priorities.
For example, Austria published its national AI strategy in 2021 and reviewed it in 2024, introducing a complementary implementation plan (Federal Chancellery of Austria, 2024[2]). Similarly, Czechia, which introduced its AI strategy in 2019, revised it in 2024 with a clear focus on harnessing AI for the benefit of the country’s economy and society (komunikace, 2024[3]). Denmark, which first published an AI strategy in 2019, introduced its 2024‑2027 Strategic Approach to Artificial Intelligence. It focuses on ethical AI, competitiveness of Danish companies and AI adoption in the public sector, with dedicated funding and new initiatives (Digitaliseringsministeriet, 2024[4]).
France, an early mover with its 2018 AI strategy, initially focused on interdisciplinary AI research and computing capacity (2018-2022). In its second phase (2022-2025), the strategy has shifted towards AI adoption across the economy – particularly among small and medium-sized enterprises (SMEs). At the same time, it seeks to enhance training, research and sovereign AI technologies, as well as incorporate new measures for generative AI development.
Ireland, following the publication of its 2021 AI strategy, introduced a revised version in 2024 to reflect major advancements in AI technology and regulation. A progress report assessed the strategy’s implementation and a high-level “refresh” introduced actions to support the effective rollout of the EU AI Act, analyse AI’s sectoral impact (including generative AI), enhance SME awareness of AI adoption benefits and expand digital upskilling initiatives (Government of Ireland, 2024[5]).
The national AI strategy in Italy, introduced in 2020, was revised in 2024 to expand AI infrastructure, align with EU priorities and introduce funding mechanisms supporting AI adoption across key sectors, including business, public administration and research. Spain also updated its 2020 AI strategy in 2024, integrating structured key performance indicators (KPIs) and monitoring mechanisms to track progress (España Digital, 2024[6]).
Several countries, namely Cyprus, Hungary, Latvia, Lithuania, Luxembourg, Malta, Portugal and Slovenia, are updating their national AI strategies. Poland, which adopted a national strategy in 2020, announced plans in November 2024 to create an AI fund and a cross-governmental AI council to oversee implementation of projects and guide AI policymaking. Sweden is redesigning its AI policies, announcing a new dedicated AI strategy to follow the more comprehensive 2025 digitalisation strategy published in May 2025.
The rise of generative AI has prompted several EU Member States to revise their strategies
The rise of generative AI has been a key driver of recent strategy updates, with about half of EU Member States indicating it prompted revisions to their strategies. Many of the 2023‑2024 revisions address generative AI capabilities, risks and opportunities. France added specific measures to its national strategy in 2023 in response to generative AI developments. These included calls for projects on generative AI building blocks and accelerating uptake, while also reinforcing AI cloud services and public supercomputing capacity. Similarly, Ireland’s 2024 strategy refresh includes actions to analyse the potential impacts of generative AI on key economic sectors. For its part, the Netherlands became one of the first EU Member States to publish a government-wide vision on generative AI in January 2024, outlining key principles such as safety, equity, human welfare and sustainability. Business Finland has also launched a campaign to support generative AI development. In Belgium, the Brussels-Capital Region introduced a programme in early 2025 that offers up to EUR 80 000 for SMEs to develop generative AI use-cases, complementing national efforts to foster adoption of generative AI.
Strategies increasingly aim to promote AI uptake across their economies and public sectors
Recent strategy updates reflect a gradual shift in emphasis towards practical AI implementation and sectoral adoption. As foundational research capabilities and policy frameworks mature, EU Member States are increasingly focusing on mechanisms to promote AI uptake across their economies and public sectors. For example, France has shifted from a research excellence agenda to one focused on economic diffusion; Denmark is emphasising public sector AI adoption; and Ireland is enhancing awareness of AI benefits among SMEs. These developments suggest that national AI policies are evolving to address implementation challenges while maintaining their commitment to research excellence and innovation. This pattern highlights the interdependent relationship between building AI capabilities and ensuring their effective deployment for economic and societal benefit.
Some AI strategies are embedded within broader plans for digital transformation
Several EU Member States, while having dedicated AI strategies, have also embedded AI within broader digital transformation frameworks. For example, after developing early AI‑specific strategies (Finland’s Age of Artificial Intelligence and Artificial Intelligence 4.0), Finland has integrated AI into its national Digital Compass, setting a long-term digital roadmap through 2030. Similarly, the national AI strategy in France is part of a massive EUR 54 billion investment plan called France 2030. It aims at strengthening French strategic sectors and positioning the country as one of the world leaders in industries that will shape the landscape of 2030. For its part, the Netherlands launched its Strategic Action Plan for Artificial Intelligence in 2019, which initially focused on AI-specific priorities. Since then, it has integrated updates emphasising innovation, education and ethical AI use into broader national digital strategies. These include the Dutch Digitalisation Strategy (2021), the Value-Driven Digitalisation Work Agenda (2022) and the Digital Economy Strategy (2023).
Croatia, Greece and the Slovak Republic were developing dedicated AI strategies at the time of writing. In 2024, Greece formalised its Blueprint for Greece’s AI Transformation. In these countries, AI policy is embedded in broader digital transformation strategies, focusing on infrastructure, research, skills development and sectoral AI applications.
EU Member States adopt varied governance models to steer AI policy and oversight
Governance structures for AI policy vary across EU Member States but generally combine centralised leadership, inter-ministerial co‑ordination and multi-stakeholder engagement (Table 2.2). Most national AI strategies are led by central government entities or ministries responsible for digitalisation, innovation, economy or research. Some countries such as Austria, Ireland and Italy have assigned leadership to the Prime Minister’s Office or Chancellery. In France, the National Coordinator for Artificial Intelligence reports to the prime minister, while in Hungary, the Prime Minister’s Cabinet Office jointly oversees AI governance together with two ministries. Other countries, like Greece, Estonia, the Netherlands, Slovenia and Spain, rely on ministries dedicated to digitalisation or innovation. In Germany, three federal ministries – responsible for education and research, economic affairs and climate action, and labour and social affairs – jointly manage AI policies. Other countries, such as Finland, through the Digital Office, involve a broad range of ministries, or rely on a more decentralised approach, as in Sweden. Finally, some countries have established specialised agencies or authorities to oversee AI implementation, such as the Malta Digital Innovation Authority, the Vinnova innovation agency in Sweden or the Portuguese INCoDe2030 skills initiative.
Inter-ministerial co‑ordination is commonly pursued through dedicated structures such as committees or working groups. Several countries reported challenges in ensuring effective collaboration and coherence across sectors. Establishing dedicated inter-ministerial task forces or assigning clear leadership to a single ministry or agency can help streamline implementation, avoid duplication and accelerate policy delivery across different sectors.
Some countries have introduced dedicated bodies to oversee AI compliance and governance. The Malta Digital Innovation Authority ensures strategic alignment with national digital transformation goals and facilitates regulatory oversight. Spain has established the Agency for the Supervision of Artificial Intelligence, responsible for AI risk assessment, regulatory enforcement and ensuring AI applications comply with national ethical and legal standards. Poland also announced plans to create the Supervisory Commission for Artificial Intelligence to oversee high-risk AI applications in sectors such as healthcare and transport.
Alongside government-led co‑ordination, many EU Member States have established multi-stakeholder engagement mechanisms to ensure that AI policies are informed by expert knowledge, industry priorities and civil society perspectives. Countries have adopted various approaches to involve a broad range of stakeholders in the formulation and revision of national AI strategies. For example, Croatia formed a dedicated working group to develop the strategy. For its part, Denmark engaged citizens through public hearings and created an all-of-government task force (central government, all five regions and 98 municipalities). Ireland conducted an electronic consultation and established a working group composed of senior officials. Malta also conducted consultations and benchmarking exercises, emphasising inter-ministerial ownership. In addition, it carried out an impact assessment, although results were not made public. Poland has established a large advisory group of over 400 experts covering various topics, including market and ethical considerations; a smaller group advises the Minister for Digital Affairs.
Many countries have further institutionalised stakeholder engagement through advisory councils, public‑private coalitions and consultation platforms that support AI policy development, implementation and monitoring. Austria launched the AI Stakeholder Forum in March 2024 as a national platform for structured engagement between government, industry associations, research institutions and civil society organisations. AI4Belgium, a coalition of public, private, academic and civil society stakeholders, acts as a hub for public-private collaboration to ensure the country’s AI policy reflects the interests of a broad range of stakeholders. In 2023, Bulgaria established the Information and Communication Technology (ICT) Community Advisory Board as an expert advisory group. It brings together public and private stakeholders to advise on legislative changes and facilitate interaction between government and industry in digital governance. Similarly, the AI Coalition in Hungary, with around 500 members from public, private and academic sectors, supports both strategy development and project implementation. Ireland has established the AI Advisory Council and the Enterprise Digital Advisory Forum, which provide expert guidance on AI trends and help policymakers adapt strategies to technological advancements and economic needs.
Monitoring national AI strategies relies on specific KPIs or EU Digital Decade targets
Effective monitoring and evaluation are essential components of national AI strategies, enabling policymakers to assess progress, ensure effective implementation and make necessary adjustments to achieve strategic goals. Across EU Member States, approaches to monitoring and evaluation vary widely. While some countries lack dedicated mechanisms, others, such as Austria, Czechia, Estonia, France, Malta, Romania, Slovenia and Spain, are actively tracking or planning to track KPIs directly tied to their AI strategies. In contrast, countries like Croatia, Ireland, Lithuania, Luxembourg, the Netherlands and Portugal monitor AI developments within broader digital transformation frameworks. To that end, they often align with EU‑level KPIs, such as those from the EU Digital Decade. Belgium and Greece are establishing AI observatories to collect and disseminate relevant data. Meanwhile, countries like Germany and Ireland have yet to define overarching KPIs, with monitoring responsibilities often distributed across ministries overseeing specific projects or initiatives.
Several EU Member States have defined KPIs for their AI strategies and perform regular monitoring and evaluation, including through regular review cycles and external assessments. In Austria, the AI Implementation Plan 2024 includes an evaluation monitor that tracks 65 measures, with 55 successfully completed to date. Malta also follows a structured approach, tracking its AI strategy through 72 specific action points, with clear implementation metrics. Similarly, Spain has systematically integrated KPIs into its AI strategy, with operational indicators tracking implementation timelines and deliverables. It reviews project-level impact indicators quarterly and submits annual implementation reports to the Council of Ministers.
Some EU Member States track AI progress within the context of broader digital transformation frameworks rather than through dedicated AI monitoring systems. Several countries refer to the AI-related indicators in the EU Digital Decade framework – notably AI adoption in businesses – as the KPI to monitor progress in AI advancement at the national level. Croatia tracks progress in AI-relevant areas as part of the Digital Croatia Strategy 2032 and the 2030 National Roadmap for Digital Decade Policy roadmap. Likewise, Finland monitors its AI policies as part of its national digitalisation strategy without dedicated KPIs but through cross-sectoral performance reviews. Latvia and Lithuania integrate AI monitoring into their national digital governance strategies, tracking AI adoption through innovation and industrial policy evaluations rather than dedicated AI-specific mechanisms. Similarly, the Netherlands evaluates AI initiatives as part of its broader digitalisation agenda, aligning monitoring efforts with the EU Digital Decade reporting framework.
Note: “..” indicates information not available.
1. In May 2025, Luxembourg launched a new national AI strategy. While not considered here for analytical purposes, it is mentioned for ease of reference (Government of Luxembourg, 2025[59]).
Source: Based on data reported by EU Member States through the survey and interviews.
More than half of EU Member States reported having evaluated their national AI strategies, either for internal use to inform policy revisions or as external assessments made publicly available. However, the scope and depth of these evaluations vary significantly. In some cases, they comprise basic monitoring of implementation status, while others involve in-depth assessments of specific components or outcomes. In France, for example, the Court of Auditors conducted an external evaluation of the research pillar of the national AI strategy (Cour des comptes, 2023[7]) in 2023. In 2023‑2024, three lead ministries in Germany jointly evaluated their national AI strategy and engaged the OECD to support the process through an international benchmarking of the German AI ecosystem (OECD, 2024[8]). Transparency in AI monitoring also varies across EU Member States. Countries such as Austria, France and Spain publish evaluation results via dedicated websites or official reports, supporting public accountability. In contrast, others primarily use evaluation outcomes internally, without making them publicly available.
AI strategies and policies mainly target AI healthcare and the public sector, while reflecting unique national contexts
EU Member States take diverse approaches to integrating AI into various sectors. A large majority (20 countries) prioritise specific industries, while others adopt horizontal, cross-sectoral frameworks. Notably, Estonia, Finland, Slovenia and Sweden fall into the latter group, opting not to target any sector in their strategies. Spain has embedded its AI strategy within its Recovery and Resilience Plan, directly linking AI development to broader economic recovery objectives. As such, it adopts a horizontal approach, although it also foresees implementation of sector-specific measures in selected areas.
National AI strategies and policies of EU Member States reveal diverse sectoral priorities, reflecting both shared goals and country-specific needs. There is a strong emphasis on sectors with high potential for public impact (Figure 2.1), with healthcare and the public sector emerging as the most consistently targeted domains. Mobility, including transport and logistics, is another frequently cited sector. This focus is aligned with broader EU goals around smart mobility and decarbonisation. Similarly, climate and the environment feature prominently, with countries aiming to leverage AI to support sustainability and energy efficiency. The emphasis on energy also underscores the potential role of AI in managing renewable resources, grid optimisation and the broader green transition in the European Union. Education is an emerging area of focus, with countries integrating AI into learning systems and workforce development. Agriculture also appears in several strategies, particularly in countries with strong farming sectors.
While these priorities are consistent with the EU Coordinated Plan on AI, national strategies also reflect country-specific economic structures and strategic interests. For example, Cyprus prioritises AI in tourism and shipping, which are key to its economy. Meanwhile, Slovenia emphasises language technologies and cultural preservation, supporting its national identity. In Greece, culture, language, tourism and sustainability are key areas of focus of the national AI Factory under development. These complement ongoing measures focused on AI and digitalisation in healthcare.
These variations highlight how EU Member States tailor their AI strategies to their socio-economic contexts, competitive advantages, innovation ecosystems and policy priorities, while still aligning with broader EU goals.
Source: Data reported by EU Member States through the survey.
Notes: Based on 27 survey responses. Four EU Member States indicated their national AI strategy or policy does not prioritise any specific sector. Two EU Member States did not respond to this question.
EU Member States vary widely in how they allocate and report public funding for AI initiatives
Initiatives to promote AI development and uptake are often part of broader digitalisation agendas. As a result, it is generally difficult for EU Member States to isolate and account for AI-specific public expenditures. However, several declared they were planning to set up mechanisms to monitor that spending category. For similar reasons, data on the budgetary endowments of national AI strategies and policies do not allow for straightforward cross-country comparison. The financial amounts presented in this report reflect estimates provided by EU Member States through the survey and interviews. In most cases, these figures represent budgetary commitments rather than actual expenditures and may include contributions from both EU funding instruments and private sector sources.
There is significant variation in how EU Member States fund AI national initiatives. Only 12 of 27 EU countries have dedicated AI strategy budgets with specific financial allocations. The remainder finance AI through specific initiatives led by relevant ministries or rely on AI funding integrated into broader digitalisation strategies (Table 2.3). All EU Member States reported multiple initiatives aligned with each thematic area of the EU Coordinated Plan on AI, financed through national budgets or EU-level programmes (see country notes for details).
Among countries with dedicated AI budgets, France and Germany reported the highest annual allocations. Germany invests between EUR 482.8-625 million annually as part of a EUR 5 billion strategy running from 2018 to 2025, with EUR 3.38 billion spent by 2024. France allocates EUR 250 million through its AI strategy, complemented by EUR 750 million from additional sources, with an estimated commitment of EUR 4 billion for 2022‑2025. Although Italy lacks a dedicated AI strategy budget, it indicated an estimated yearly investment of EUR 270 million. This will be complemented by other funding sources targeting broader digitalisation efforts, such as tax credits for enterprises and equity investments, which include AI-driven advancements. Czechia allocates funding for AI through specific initiatives outlined in the Action Plan for the Implementation of Digital Czechia. It has an overall budget of approximately EUR 753.7 million (CZK 19 billion) designated for all active projects in 2025. Mid-range investments come from countries like Spain (EUR 100 million), Estonia (EUR 31.7 million), Slovenia (EUR 22.5 million) and Bulgaria (EUR 5‑20 million). Meanwhile, Cyprus, Denmark and Malta reported allocations of around EUR 2.5‑5 million annually.
Regional variations within countries also emerge in the data. Belgium, for instance, demonstrates strong regional investment through its Flanders region. It allocated EUR 35 million for AI policy and EUR 118.5 million for AI research and innovation (R&I) in 2024, despite having no dedicated national AI budget. From the R&I budget of Wallonia, 4% has been allocated to AI (on average between 2022 and 2024).
Priorities for different dimensions of the EU Coordinated Plan on AI and use of EU funding sources, such as the RRF, also vary significantly among EU Member States. AI-related projects within Horizon 2020 and Horizon Europe (ECA, 2024[9]), as well as within the RRF, do not have specific tags. Consequently, it is challenging to have a complete overview of financing through EU programmes.
Note: RRF funding reported in this table only concerns the overall strategies. RRF funds for individual areas have been reported separately.
Source: Based on data reported by EU Member States through the survey and interviews.
Leveraging the RRF to advance AI initiatives
The RRF is the primary instrument for the European Union to provide financial support to Member States in the aftermath of the Coronavirus disease 2019 (COVID-19) pandemic (OECD, 2024[10]; EC, 2025[11]). It has a total allocation of EUR 672.5 billion – comprising EUR 312.5 billion in grants and EUR 360 billion in loans. In this way, the RRF aims to foster economic recovery while advancing the EU green and digital transitions.
Under the National Recovery and Resilience Plans (NRRPs), EU Member States had to indicate objectives and funding aimed at accelerating the green and digital transitions. The national AI strategy in Spain is a core element (Component 16) of the digital axis of the country’s NRRP and identified as a reform to be launched as part of the plan. Lithuania and Romania indicated the instrumental role of the RRF and other EU funding for AI-specific publicly funded programmes. Meanwhile, Slovenia reported that about half of planned investments in the national strategy (i.e. approximately EUR 65 million) will be funded through the RRF.
EU Member States have leveraged RRF resources to support a wide range of initiatives. Notably, these include the areas of enabling conditions for AI development and uptake; fostering excellence from laboratory to market; ensuring AI technologies work for people; and building strategic leadership in high-impact sectors.
A substantial number of EU Member States have used RRF funding to set enabling conditions for AI development and uptake. These initiatives frequently focus on improving data governance, enhancing interoperability, expanding cloud infrastructure and developing national data platforms. In Belgium, for instance, Smart Data Sciences Market Consultation and SolidLab Flanders initiatives enhance data sharing, interoperability and citizen control over personal data. Latvia and Lithuania have invested in secure government cloud environments and open data platforms to foster data accessibility and trust. Spain made use of the funding to create sectoral data spaces and enhance HPC capacities. Similarly, countries like Poland, Romania and Slovenia have directed funding towards creating digital public infrastructure and improving data-management capabilities.
In parallel, several EU Member States have deployed RRF funding to support the transition from R&I to market-ready AI applications, particularly by targeting SMEs and innovation ecosystems. These efforts align with the ambition to make the European Union the right place for AI excellence. For example, the EDIH CROBOHUB++ digital innovation hub in Croatia provides tailored support to SMEs in adopting digital solutions. In Greece, the Digital Transformation of SMEs initiative delivers training, guidance and funding to modernise business processes through AI and related technologies. Denmark and Ireland have similarly focused on enhancing SME competitiveness by facilitating access to digital tools and resources, as well as building national networks for testing and experimentation.
Another prominent area of investment through the RRF involves ensuring that AI technologies work for people, with an emphasis on digital skills development and inclusive access to training. This thematic strand reflects the understanding that a thriving AI ecosystem depends on a well-prepared workforce. The adult learning platform in Bulgaria exemplifies this approach, offering personalised digital learning pathways supported by AI to upskill workers across sectors. Italy and Portugal have platforms that guide users towards reskilling and employment opportunities, often targeting vulnerable groups or those affected by labour market transitions. The AI Research network in Ireland and efforts in Lithuania to integrate AI into public services also illustrate how countries are embedding human-centred approaches in national strategies.
Finally, several EU Member States have employed RRF funding to build strategic leadership in priority sectors, such as healthcare, education, justice and mobility. These initiatives reflect national efforts to harness AI for public value creation and sectoral modernisation. In Spain, the national AI strategy includes targeted investments in green data centres and sovereign cloud infrastructure to support long-term digital sovereignty. Finland and Slovenia have focused on using AI to improve public administration, while Croatia and Romania are using AI to strengthen social protection systems and job market platforms. In Belgium, the regional government in Flanders has made notable investments in AI R&I innovation, demonstrating how subnational actors are also contributing to strategic leadership in this domain.
Note: Italy also reported EUR 30 billion in the area of “Set enabling conditions for AI development and uptake in the European Union”.
Source: Data reported by EU Member States through the survey and interviews.
Harnessing the potential of data is important for the economy and to promote fairness, inclusion and diversity in society. Access to quality data and ability to collect, store and manage those data are essential components of AI ecosystems. Quality data requirements are particularly significant when it comes to developing foundation models such as large language models. As noted by the 2021 EU Coordinated Plan on AI, AI has growing influence in both the economy and society. Consequently, the potential of data should be harnessed both from an economic and industrial perspective, and from the standpoint of fairness, inclusion and diversity.
Addressing the incomplete nature of language data is crucial in this context. Without developing language models beyond English, increased use of generative AI may prevent large fractions of the population from reaping the benefits of AI (Ta and Turner Lee, 2023[12]). About 56% of open-source datasets from Hugging Face, an online platform for foundational model hosting and sharing, are in English (OECD, 2024[10]). Similarly, by one estimate, about 93% of large language model GPT-3 training data come from sources in English. Moreover, trusted and high-quality data are necessary for deploying AI systems in sensitive or sovereignty-related areas such as defence or energy (French Government, 2024[13]).
In parallel, the infrastructure required to develop, train and deploy AI systems – particularly at scale – has seen exponential growth in demand. Hyperscale data centres have emerged as essential infrastructure, offering massive computing power, high-speed networking and expansive data storage. The global hyperscale infrastructure market is dominated by a few major players – Amazon Web Services, Google Cloud Platform, Microsoft Azure and Alibaba Cloud. Indeed, the first three account for roughly two-thirds of the global public cloud market.
As AI workloads expand, so too does the need for AI-optimised infrastructure. Current projections suggest that demand for data centre capacity could triple by 2030, growing between 19% and 27% annually. By that time, generative AI alone is expected to account for approximately 40% of total AI-related infrastructure demand. Meeting this surge will require building at least twice the data centre capacity constructed globally since the year 2000 (McKinsey & Company, 2024[14]; BBC, 2025[15]).
Computing capacity – referred to as AI compute – is another cornerstone of AI capability and competitiveness. In a list of the 500 most powerful non-distributed computer systems, the European Union hosts only three EU-based supercomputers in the top ten. The United States dominates the list with five entries, including the top four. The remainder are in Japan and Switzerland (one each) (Top 500, 2025[16]). EU Member States host 132 supercomputers (in 19 countries). Germany leads the ranks with 38, followed by France and Italy, with 24 and 14, respectively.
The European Union also lags in market share for semiconductors, which could make it vulnerable to market shifts. Integrated circuits or computer chips made of semiconductors are the foundational hardware for AI compute. Yet, the global semiconductor supply chain is marked by high levels of market concentration and geographic asymmetry. Indeed, Nvidia’s model CUDA, based in the United States, has become a de facto standard for graphic processing unit acceleration in AI. This makes the industry vulnerable to both economic shocks and geopolitical tensions.
The chip manufacturing market presents a similar landscape. In the advanced chip segment (3‑7 nanometres), Chinese Taipei-based TSMC has a market share of about 90%. The other main actor is Samsung in Korea. ASML, the world’s leading manufacturer of advanced chipmaking machines (using extreme ultraviolet lithography systems), is based in the Netherlands.
To mitigate related economic and geopolitical risks, major world economies have taken steps to bolster their chip manufacturing capacity (France Digitale, 2024[17]). In the European Union, key measures include the Chips Act and support measures from Member States. France, for example, has provided EUR 7.4 billion worth of direct grants to GlobalFoundries and STMicroelectronics to develop a large-scale manufacturing site for high-performance chips (EC, 2023[18]).1 In December 2024, the European Commission announced the establishment of seven AI Factories (a EUR 1.5 billion investment combining national and EU funding). This is expected to double EuroHPC computing capacity and thereby help enhance European AI capability (EC, 2024[19]). The following month, complementing the CHIPS for America fund, the United States announced USD 500 billion worth of investments to roll out AI infrastructure, particularly data centres and the attendant power supply (Politico, 2025[20]).
To help harness the potential of data, the EU Coordinated Plan on AI calls upon Member States to:
Invest in strengthening Europe’s position in next-generation cloud and edge technologies and foster cloud uptake through their national recovery and resilience plans and in line with the example component for the RRF ‘Scale-up’ flagship 64 and including through multi-country projects.
Investing in advanced cloud computing infrastructure will be required to ensure the affordability and availability of computing resources by a broad range of economic actors for AI-related purposes and the wider digital transformation. Edge computing, in turn, can help reduce energy consumption, reduce latency, increase the performance of high bandwidth applications and enable new cloud service models. Edge computing could also support resilience, security and privacy protection (OECD, 2022[21]).
The EU Coordinated Plan on AI also outlines a series of actions to enhance data processing and compute capacity at the national, regional and EU levels. Specifically, it calls for further developing the necessary technological systems and AI-enabling infrastructure, fostering access by all relevant AI actors to compute resources and expertise, and strengthening the semiconductor ecosystem.
To foster critical compute capacity, Member States are encouraged to pursue the following actions:
Continue the development of national integrated large-scale data management and HPC infrastructure to support research, innovation and skills development in AI through regional, national and European digital innovation hubs.
Ensure that academic, industry and public sector organisations can leverage national HPC and data-management infrastructure and expertise to optimise and scale up their AI innovation and applications.
Invest in strengthening the EU position in processors and semiconductor technologies for AI through their NRRPs, in line with the example component for the RRF scale‑up flagship area and through multi-country projects.
This section discusses actions by EU Member States to tap into the potential of data and foster critical computing capacity, in line with the EU Coordinated Plan on AI. Table 2.4 summarises key findings from the survey and complementary interviews.
EU Member States are strengthening data foundations for AI
Across the European Union, countries are developing a wide range of initiatives to tap into the value of data. While these efforts are shaped by national priorities, they often share the common goals of improving access to and sharing of data, modernising infrastructure and supporting digital innovation. In total, EU Member States reported 57 initiatives (Figure 2.3). Several governments have launched overarching strategies to guide how data are used, shared and protected across public and private sectors. Some countries are focusing on secure infrastructure like data centres, while others are working on creating environments where data can be easily shared and re-used, including through open data policies and data spaces. A growing number of EU Member States also reported specific initiatives to enhance availability of datasets in national languages (see Chapter 3).
Note: Data for Belgium include regional initiatives.
Source: Data reported by EU Member States through the survey and interviews.
More than half of EU Member States reported a data strategy or action plan, or ongoing efforts to develop one
Fifteen EU Member States have adopted national strategies to ensure data are managed consistently and responsibly. These strategies reflect a shared understanding of data as both a public good and a strategic asset to drive economic growth, requiring careful governance to balance innovation with protection of fundamental rights. In most cases, they also position themselves within the broader EU data framework. Austria published its first national data strategy in October 2024 (Federal Chancellery of Austria, 2024[22]). The strategy, whose development relied on broad stakeholder involvement, encompasses areas such as sustainable infrastructure, responsible data sharing and innovative data culture/data literacy. It is also aligned with the EU data strategy (EC, 2020[23]). In Germany, the National Data Strategy aims to maximise the potential of data across sectors, including public administration, research, industry and civil society. Key objectives include expanding data availability by creating more datasets and facilitating access to government data, improving data quality through standardised descriptions and quality assurance, and fostering a culture of data-driven decision making across society. These efforts are largely driven by the development of the National Research Data Infrastructure (German Federal Government, 2023[24]) The Greek National Data Strategy, part of the Digital Transformation Bible 2020‑2025, aims to create a balanced data ecosystem that respects individual privacy, promotes open government and data-driven innovation, and complies with EU obligations and international best practices. In Portugal, the National Data Strategy envisions a governance model that fosters a collaborative and dynamic ecosystem, promoting data altruism and creating synergies among stakeholders (Portugal Digital, 2022[25]).
Several countries integrate data strategies within broader digital agendas. In Germany, Digital Strategy 2025 consolidates priorities for digital transformation under a unified framework. In Denmark, the National Strategy for Digitalisation includes improved data access as part of “a strong, ethical and responsible digital foundation”. In Finland, Digital Compass embraces data policy within its broader digitalisation vision. In Sweden, the strategy is linked to innovation in AI.
Some data strategies or action plans of EU Member States focus on public administration. Czechia adopted a Strategy for Data Management in the Public Administration. It focuses on enhanced data sharing among public administrations through standardisation to modernise services and optimise efficiency through improved data management. France has several data-related strategic initiatives for the public sector, both cross-cutting (inter-ministerial data administration framework) and domain-specific (e.g. relationships between the public administration and citizens). In addition to open government data initiatives and laws, the 2021 Inter-ministerial Framework for Data Administration in France covers data, algorithms and source codes (Ministère de la transformation et de la fonction publiques, 2021[26]). In Ireland, the Public Service Data Strategy for the period 2019‑2023 (Government of Ireland, 2019[27]) sets out goals and actions to deliver a more joined‑up approach to managing and using data in the public service; a new strategy was expected in 2025. Similarly, Italy focuses on improving how government data are managed and used to support services. Meanwhile, the Public Administration Data Strategy 2023‑2027 in Malta aims to establish single, centralised registers for individuals, organisations, geospatial data and government administrative data. Other countries, like Latvia, Lithuania and Luxembourg, are still developing or updating their strategies, working on building new governance models to support the use of data across government systems.
There are also data strategies at the subnational level. In 2022, the association of Danish municipalities published its data strategy for local governments (KL, 2022[28]), which notably calls upon municipalities to make standardised data available. In Belgium, the Flemish Data Strategy (for which an updated version is being developed) promotes open data availability, standardisation and interoperability. Moreover, it fosters collaboration between local and regional authorities to enable data-driven decision making.
Governments are making more public data available to support transparency and innovation
Open data initiatives represent a key area of focus across multiple EU countries, with varying approaches to increase transparency, accountability and data re-use. Croatia (Open Data Portal), Denmark (datavejviser.dk) and Poland (Dane.gov.pl) have established centralised repositories where public authorities can publish datasets, creating single access points for users. Legal frameworks vary – from the 2016 Law for a Digital Republic in France that lays the foundation for open public data, to the Act on Open Data and the Re-use of Public Sector Information in Poland. The Open Government Action Plan 2023‑2027 in Denmark supports development of data policies aligned with principles of open government, emphasising transparency, accountability and citizen participation. Similarly, the Polish Public Open Data Programme aims at enhancing the accessibility and usability of public sector information. Some countries have developed specialised repositories. The Portal for Open Science in Bulgaria, for example, has more than 73 000 science-related documents available for access. Meanwhile, France has progressively opened court decision data beginning with Supreme Court decisions in 2021.
Data spaces are emerging as shared environments for secure and structured data exchange
A growing number of countries are developing shared data environments to enable secure and standardised data exchange across sectors. In Austria, the Data Intelligence Offensive (DIO) aims to promote business models for data exchange and monetisation. It supports the development of data spaces in various sectors, including tourism (Tourism Data Space, 2024[29]), health (DIO, 2024[30]) and agriculture. Moreover, the DIO supports the regional data space.tirol (digital.TIROL, 2024[31]). Flanders region in Belgium hosts initiatives such as Athumi (2024[32]) and SolidLab (2024[33]), which build secure and privacy-respecting data ecosystems. Led by GATE, the Urban Data Space is the first sectoral data space in Bulgaria. It facilitates data sharing and access between organisations to enable development of new business products and services. The Netherlands established the Centre of Excellence for Data Sharing and Cloud to support development of data spaces for organisations to securely exchange data across sectors, such as the Smart Connected Supplier Network (2024[34]) and HDN (2024[35])2. In Spain, whose national AI strategy already encompasses several data‑related initiatives, an RRF-supported Plan to Promote Sectoral Data Spaces was released in November 2024 (Government of Spain, 2024[36]). Lithuania and Sweden are focusing on integrating data lakes and open platforms into digital public services. Germany and Greece reported to be actively supporting the federated data infrastructure Gaia‑X (2024[37]), which has national hubs in most EU Member States.
Countries are building the infrastructure to manage and store growing volumes of data
Many EU Member States are expanding their physical and digital infrastructure to support the increasing demand for secure, scalable data processing and storage. Some EU Member States mentioned the development of data centres to support the digitalisation of government services (i.e. beyond AI). Italy has invested in large-scale national data centres that underpin government services and digital transformation (see also below on cloud infrastructure). With assistance from the Digital Europe Programme, Czechia is setting up a shared data pool to facilitate smoother and more efficient data sharing and collaboration across public administrations. In a similar vein, the Shared Government Data Centre in Ireland aims at supporting digitalisation of government services (Government of Ireland, 2021[38]). Slovenia and Sweden are strengthening research infrastructure through data storage platforms tailored for science and higher education. These facilities often form part of broader strategies that also include national cloud systems and supercomputing hubs. For instance, Portugal and Slovenia are integrating these investments with HPC and edge capabilities.
EU Member States are implementing national cloud initiatives to support development of secure and interoperable cloud infrastructure
About two-thirds of EU Member States reported having cloud strategies or initiatives in place (Table 2.5). While the scope and scale of these initiatives vary, they share the common goals of facilitating efficient and secure cloud adoption, particularly within public administration. At the same time, they promote digital sovereignty, interoperability and innovation. Several EU Member States also reported alignment with EU-level initiatives such as Gaia-X and the European Open Science Cloud.
Most of these initiatives prioritise development of cloud infrastructure for the public sector, often guided by “Cloud‑First” policies. For instance, Finland updated its central government cloud guidelines in 2023 to reinforce a Cloud‑First approach. Denmark operates GovCloud, a dedicated cloud service for government entities. For its part, Greece has invested in the Government Cloud (G-Cloud Services) and the Hybrid Cloud project. The Flemish Administration of Belgium also demonstrates this focus with its Cloud Strategy.
Several countries have adopted comprehensive national strategies, supported by substantial public and EU funding. France has launched an ambitious EUR 1.8 billion industrial plan under its National Cloud Strategy to strengthen its domestic cloud ecosystem. This includes the PEPR Cloud research programme, initiated in 2024 to advance innovation in areas such as security, sustainability and performance. The Administrative Cloud Strategy in Germany aims to build a modular, interoperable federal cloud infrastructure through open standards, reduced vendor dependency and shared platforms. Italy has established a Cloud Strategy and National Strategic Hub with supporting data centres. Portugal has a Public Administration Cloud Strategy alongside initiatives to attract further data centre investments. Romania has committed over EUR 2.7 billion through its NRRP to support the Deployment of the Government Cloud Infrastructure measure and cloud migration efforts.
Other countries such as Poland, Portugal and the Slovak Republic have adopted detailed frameworks that include cloud and edge computing components. These frameworks typically encompass implementation guidelines, training programmes, monitoring mechanisms and streamlined procurement processes.
Other countries are implementing hybrid and federated cloud infrastructures, leveraging both national and external capabilities. Malta is rolling out a Hybrid Cloud Enabling Infrastructure, while Lithuania is reforming ICT management. The latter includes hybrid cloud deployment; integration of state information resources into a state data lake; and development of a national cloud infrastructure. Bulgaria is also advancing its hybrid cloud capabilities. Luxembourg, whose AI strategy is closely tied to its cloud development goals, has introduced Clarence, a sovereign cloud platform.
The Important Project of Common European Interest (IPCEI) on Next-Generation Cloud Infrastructure and Services (CIS) was announced on 5 December 2023. ICPEI-CIS authorised seven Member States (France, Germany, Hungary, Italy, the Netherlands, Poland and Spain) to mobilise up to EUR 1.2 billion in public funding to support research, development and initial industrial deployment of European innovations in cloud and edge technologies (EC, 2023[39]). This project aims to establish a decentralised cloud-edge ecosystem that will reduce technological dependencies and foster new data-driven business models. Its primary goal is to create a Multi-Provider Cloud-Edge Continuum, which allows for advanced use of data processing resources from cloud to edge environments. This open ecosystem is expected to support AI and Internet of Things applications across various industries, including manufacturing, mobility and energy (BMWK, 2023[40]; 8ra Cloud-Edge Continuum, 2024[41]). Most initiatives reported by EU Member States regarding next‑generation cloud and edge technologies concern this project, which is supported by RRF funding.
Source: Data reported by EU Member States through the survey and interviews.
Strengthening compute infrastructure that can support AI workloads is a key objective in most EU Member States
More than two‑thirds of EU Member States (23) reported initiatives to strengthen compute infrastructure (Table 2.6). Several of these initiatives are connected to the European High Performance Computing Joint Undertaking (EuroHPC JU) – a key structuring element in shaping the EU supercomputing landscape. Notably, two supercomputers in the top ten of the TOP500 list – LUMI (Finland) and Leonardo (Italy) – have been procured through the EuroHPC JU (EuroHPC, 2025[42]). This structuring role is bound to remain significant. The launch of the AI Factories (EC, 2025[43]) will bring together supercomputers, data and human capital to develop an ecosystem to train advanced AI models and develop AI solutions. The EuroHPC JU will provide the necessary supercomputing infrastructure, covering half of the acquisition and operation costs of AI-optimised supercomputers, as well as half the cost of the services provided by AI Factories (Neistadt, 2025[44]).
HPC resources can support workloads both specific and non-specific to AI for a variety of scientific research and industrial applications. However, AI-related compute needs appear to be an increasingly important driver. Several EU Member States reported to have stepped up investment to enhance compute capacity in 2023‑2024. This move responded to the needs associated with the latest developments in AI technology such as large language models. It also reflected the soaring demand for AI compute, especially for deep neural networks (OECD, 2024[10]). Italy, which hosts 5 of the world’s 500 most powerful supercomputers, continues to bolster its infrastructure. Meanwhile, Poland is preparing to launch two new facilities, the AI Factory at CYFRONET and the PIAST AI Factory under the EuroHPC.
EU Member States are also becoming increasingly aware of the need to make sufficient compute capacity available to all relevant stakeholders in the AI ecosystem – not only researchers but also the public administration and private companies, including SMEs. For instance, France deployed the supercomputer Jean Zay during the first phase of the country’s AI strategy between 2018 and 2022. In 2023, France announced it would expand the capacity of Jean Zay in response to generative AI needs and improve access for AI start-ups. The associated teams of support engineers have also been further strengthened. Similarly, in Luxembourg, the MeluXina supercomputer supports AI, industry and research use-cases and forms part of the broader national HPC and quantum strategy. The Komondor HPC and OTP-Emese systems in Hungary serve both academic and commercial purposes, including AI model training.
Given the sizeable investment outlays required, initiatives to enhance compute infrastructure often rely on collaboration across several partners:
The Gauss Centre for Supercomputing in Germany brings together the country’s three leading national supercomputing centres in Jülich, Leibniz and Stuttgart. The centre, which is instrumental in advancing Germany’s capabilities in computational science and engineering, is jointly funded by the German federal government and the respective host federal states (GCS, 2024[45]).
The Vienna Scientific Cluster in Austria is another collaborative effort, pooling resources from several universities and providing supercomputer resources and technical support to their users (VSC, 2024[46]). Moreover, with the help of RRF resources, the Quantum Austria funding initiative will enable development of the country’s fastest HPC system: the Multi-Site Computer Austria (MUSICA) (FFG, 2024[47]).
The EuroQCS-Poland project, in turn, is a joint initiative of Polish and European institutions, owned by the EuroHPC JU and located at the Poznan Supercomputing and Networking Center. EuroQCS-Poland will be a digital, gate-based quantum computer, based on trapped ions offering 20-plus physical qubits. The infrastructure will allow for remote access (EuroHPC JU, 2024[48]).
Belgium, Croatia and Finland are also actively participating in EuroHPC JU initiatives, including LUMI-Q and IPCEI-CIS.
Bulgaria reported several initiatives aimed at enhancing HPC capacity. These include Discoverer (launched in 2021), a petascale supercomputer co-funded through the EuropHPC JU and specifically designed to support big data processing and AI applications; supercomputer HEMUS (launched in 2023); and supercomputer Avitohol (expected to be operational in 2025).
The HPC roadmap in Slovenia includes supercomputing and edge computing investments through projects DRO Next and SLING.
Sweden is complementing investments in AI-focused HPC by participating in the LUMI supercomputer hosted in Finland and the planned deployment of its own national systems by 2025.
Public-private partnerships are also emerging. Denmark started collaborating with leading AI hardware and software supplier NVIDIA in 2024. Together, they plan to establish a centre for AI innovation that will host one of the world’s most powerful AI supercomputers (Gefion). This public-private partnership is co-funded by the Export and Investment Fund in Denmark and the Novo Nordisk Foundation. It is expected to provide researchers from the Danish public and private sectors with access to a state-of-the-art supercomputer. It will be optimised for large‑scale projects using AI, as well as software platforms, training and expertise (Novo Nordisk Foundation, 2024[49]).
Several EU Member States are developing HPC and quantum compute capacities as part of a national or regional ecosystem. In some cases, such as Spain, initiatives focusing on quantum computing are embedded into national AI strategies. Conversely, countries like Austria and Denmark have opted for separate programmes or strategies. In Finland, the government supports quantum research through initiatives like the LUMI-Q project. It has committed funding to scale its quantum computer to 300 quantum bits by 2027. The Quantum 2030 strategy in Ireland outlines a comprehensive framework around research, talent and collaboration. In Italy, the National Quantum Science and Technology Institute brings together public research institutions to accelerate quantum R&D. Similarly, Portugal launched the Portuguese Quantum Institute and is developing a broader strategy on advanced computing and quantum technologies, including new national facilities. The Slovak Republic is also building national capabilities in quantum communication and infrastructure. In 2023, Spain announced several measures in this area, notably to reinforce the capacity of the Barcelona Supercomputing Centre (designated in 2024 among the EU AI Factories). It plans to make such capacity available to industry actors and reinforce the Spanish HPC network. The same year, Spanish authorities announced that the first quantum computer in southern Europe would be in Barcelona (Presidencia del Gobierno, 2023[50]). Moreover, the Spanish government was developing at the time of writing a National Strategy on Quantum Technologies3. This builds on the Quantum Spain programme, which is part of the country’s AI strategy to develop competitive and comprehensive quantum computing infrastructure (Quantum Spain, 2025[51]).
Note: The survey did not include a specific question on quantum initiatives; therefore, the initiatives reported reflect Member States’ own decisions to provide this information voluntarily.
Source: Data reported by EU Member States through the survey and interviews.
As a complement to their AI strategies, several EU Member States have started to support investment in the semiconductor ecosystem
More than half of EU Member States reported initiatives to support investment in the semiconductor ecosystem, often with sizeable funding. EU Member States also reported their participation in the IPCEI on Microelectronics and Communication Technologies (ME/CT) (Box 2.1) and in the Chips Joint Undertaking (EU, 2023[52]), aimed at boosting development and adoption of advanced nano‑electronic chip technologies and systems manufactured in Europe.
Several EU Member States have launched national strategies to support semiconductor development
Across the European Union, a growing number of EU Member States have adopted or are adopting national strategies to strengthen their domestic semiconductor sectors. This move reflects growing awareness of the strategic importance of the semiconductor supply chain and the need to reinforce strategic autonomy and resilience in line with the EU Chips Act. France, for example, committed to mobilise EUR 5 billion through the 2022 national strategy on electronics (French Government, 2022[53]). Through the PERTE Chip (Government of Spain, 2022[54]), Spain will mobilise EUR 12.25 billion earmarked for 2022‑2027 to support conception, design, production of chips and manufacturing of electronic products. Czechia launched its national semiconductor strategy (EUR 45 million) in 2024, targeting a threefold growth in the domestic semiconductor sector by 2029, with a focus on research, export and talent. Through a national strategy, Poland aims to support strategic semiconductor investment throughout the entire value chain (Government of Poland, 2023[55]; KIGEiT, 2023[56]) The 2024 National Strategy for Semiconductors in Portugal is similarly focused on boosting national scientific and technological advancements to contribute to EU resilience. Finland will launch a national research, development and innovation environment, including for semiconductors, with EUR 79 million from the 2024‑2027 budget. Ireland is also developing a national strategy following a public consultation in 2023. Meanwhile, Romania has launched a National Platform for Semiconductor Technologies, with approximately EUR 130 million allocated to enhance cross-sector collaboration and national design and manufacturing capacity. Slovenia is also finalising its national semiconductor strategy (2024‑2030) to co‑ordinate actions across ministries and stakeholders.
Many countries are participating in joint EU efforts such as the IPCEI on Microelectronics and Communication Technologies
Participation in the IPCEI ME/CT is a central mechanism through which many countries are mobilising national and EU funding. The Slovak Republic, for example, is supporting four Slovak companies selected under the IPCEI ME/CT, allocating EUR 200 million. Slovenia is supporting companies involved in the IPCEI through both annual funding (EUR 1.2 million) and project-based co-financing (EUR 1.5 million). In Malta, the Semiconductor Competence Centre is a strategic initiative spearheaded by Malta Enterprise in collaboration with STMicroelectronics in the context of the IPCEI ME/CT framework. It focuses on developing the workforce; conducting research; promoting industry-academia collaboration; attracting foreign investment; and assisting start-ups and SMEs (Malta Enterprise, 2024[57]).
Strategic investments are supporting R&D, infrastructure and industrial capacity
Many countries are combining strategic planning with concrete investments in semiconductor-related research and development (R&D) and manufacturing infrastructure. The Framework Programme for Research and Innovation (2021‑2024) in Germany includes EUR 400 million in national support (plus significant private investment). Belgium plays a central role through Imec, a world class semiconductor research centre that participates in PREVAIL, an EU initiative to develop AI hardware. Greece has committed EUR 7.8 million towards its participation in the EU Chips Joint Undertaking (Chips JU). Latvia signed a memorandum of understanding in 2022 with 12 partners and planned to join the Chips JU in 2025, backed by approximately EUR 745 000 per year in national co-funding. Sweden is promoting MyFab, the national cleanroom and nano-fabrication infrastructure. Vinnova, the national innovation agency, is co-financing up to EUR 2 million per year to support Swedish participation in the Chips JU. Romania is advancing a project dedicated to cross-border and multi-country collaboration (EUR 400 million in funding expected). Slovenia has also launched Čip.si (2025) to support SMEs and start-ups with access to R&D infrastructure and pilot lines.
Countries are also investing in skills, talent and academic-industry collaboration
Several countries have adopted measures focused on the talent pipeline, industry-academia collaboration and ecosystem support. In Spain, the Chip Chairs initiative is creating 17 university chairs to strengthen academic research and workforce development in the field of microelectronics. In Finland, Chips from the North provides EUR 10 million for training, co‑ordination and local ecosystem building in the Oulu region. In Slovenia, the Čip.si initiative also covers talent support, along with infrastructure and company support.
The first Important Project of Common European Interest (IPCEI) on Microelectronics, approved by the European Commission in December 2018, brought together 32 companies from five Member States (Austria, France, Germany, Italy, the United Kingdom) to support research and innovation in microelectronics. With public funding of up to EUR 1.9 billion, the initiative targeted the development of advanced technologies and components such as chips, sensors and integrated circuits for use in diverse applications – from consumer electronics to electric mobility systems.
In June 2023, the European Commission approved a second IPCEI, focused on Microelectronics and Communication Technologies (IPCEI ME/CT). It involves 14 Member States (Austria, Czechia, Finland, France, Germany, Greece, Ireland, Italy, Malta, the Netherlands, Poland, Romania, the Slovak Republic and Spain) and 68 projects from 56 companies.
The project’s overall objective is to enable the digital and green transformation by:
creating innovative microelectronics and communication solutions
developing energy-efficient and resource-saving electronics systems and manufacturing methods
The IPCEI will contribute to the technological advancement of many sectors, including communications (fifth and sixth generation of cellular network technology, 5G and 6G), autonomous driving, AI and quantum computing. They will also support companies active in the energy generation, distribution and use in their green transition.
The 14 Member States are expected to provide up to EUR 8.1 billion in funding, which is expected to unlock additional EUR 13.7 billion in private investments.
Source: EC (2023[58]), Approved IPCEIs in the Microelectronics Value Chain, https://competition-policy.ec.europa.eu/state-aid/ipcei/approved-ipceis/microelectronics-value-chain_en.
References
[41] 8ra Cloud-Edge Continuum (2024), “The future of Europe’s digital infrastructure”, https://www.8ra.com/ (accessed on 18 January 2025).
[32] Athumi (2024), “Trust in data collaboration”, https://athumi.eu/en/ (accessed on 3 December 2024).
[15] BBC (2025), “Tech giants are putting $500bn into ’Stargate’ to build up AI in US”, BBC, https://www.bbc.com/news/articles/cy4m84d2xz2o (accessed on 18 January 2025).
[40] BMWK (2023), “IPCEI next generation cloud infrastructure and services”, https://www.bmwk.de/Redaktion/EN/Artikel/Industry/ipcei-cis.html (accessed on 14 November 2024).
[7] Cour des comptes (2023), “La stratégie nationale de recherche en intelligence artificielle – Une stratégie à structurer et à pérenniser”, Cour des comptes, Paris, https://www.ccomptes.fr/sites/default/files/2023-10/20230403-strategie-nationale-recherche-intelligence-artificielle.pdf.
[31] digital.TIROL (2024), “digital.TIROL”, https://www.datahub.tirol/ (accessed on 27 November 2024).
[4] Digitaliseringsministeriet (2024), Strategic Approach to Artificial Intelligence, Government of Denmark, https://www.english.digmin.dk/Media/638719220318136690/Stategic%20Approach%20to%20Artificial%20Intelligence.pdf.
[30] DIO (2024), “Health data space”, Data Intelligence Offensive, https://dataintelligence.at/dataspaces/ds-health/ (accessed on 15 January 2025).
[43] EC (2025), “AI factories”, European Commission, https://digital-strategy.ec.europa.eu/en/policies/ai-factories (accessed on 25 April 2025).
[1] EC (2025), “Europe’s digital decade”, European Commission, https://digital-strategy.ec.europa.eu/en/policies/europes-digital-decade (accessed on 25 April 2025).
[11] EC (2025), “The recovery and resilience facility”, https://commission.europa.eu/business-economy-euro/economic-recovery/recovery-and-resilience-facility_en (accessed on 18 April 2025).
[19] EC (2024), “Seven consortia selected to establish AI factories which will boost AI innovation in the EU”, https://ec.europa.eu/commission/presscorner/detail/en/ip_24_6302 (accessed on 15 January 2025).
[58] EC (2023), “Approved IPCEIs in the microelectronics value chain”, European Commission, https://competition-policy.ec.europa.eu/state-aid/ipcei/approved-ipceis/microelectronics-value-chain_en (accessed on 23 February 2025).
[39] EC (2023), “IPCEI on next-generation cloud infrastructure and services to boost Europe’s digital decade”, European Commission, https://digital-strategy.ec.europa.eu/en/news/ipcei-next-generation-cloud-infrastructure-and-services-boost-europes-digital-decade (accessed on 18 January 2025).
[18] EC (2023), “State aid: Commission approves French measure to support STMicroelectronics and GlobalFoundries to set up new microchips plant”, European Commission, https://ec.europa.eu/commission/presscorner/detail/en/ip_23_2447 (accessed on 18 January 2025).
[23] EC (2020), “A European strategy for data”, European Commission, https://digital-strategy.ec.europa.eu/en/policies/strategy-data (accessed on 15 January 2025).
[9] ECA (2024), “EU artificial intelligence ambition: Stronger governance and increased, more focused”, Special Report, European Court of Auditors, Luxembourg, https://doi.org/10.2865/923687.
[6] España Digital (2024), Estratregia de Inteligencia Artificial 2024, [English translation], España Digital, https://espanadigital.gob.es/estrategia-de-inteligencia-artificial-2024 (accessed on 22 January 2025).
[52] EU (2023), “Chips joint undertaking”, European Union, https://european-union.europa.eu/institutions-law-budget/institutions-and-bodies/search-all-eu-institutions-and-bodies/chips-joint-undertaking_en (accessed on 22 March 2025).
[42] EuroHPC (2025), “The European high performance computing joint undertaking (EuroHPC JU)”, https://eurohpc-ju.europa.eu/supercomputers/our-supercomputers_en (accessed on 24 April 2025).
[48] EuroHPC JU (2024), “Advancing European quantum computing: Signature of the procurement contract for EuroQCS-Poland”, European High Performance Computing Joint Undertaking, https://eurohpc-ju.europa.eu/advancing-european-quantum-computing-signature-procurement-contract-euroqcs-poland-2024-07-10_en (accessed on 18 January 2025).
[22] Federal Chancellery of Austria (2024), “Datenstrategie für Österreich”, [English translation], https://www.digitalaustria.gv.at/Strategien/Datenstrategie.html (accessed on 22 January 2025).
[2] Federal Chancellery of Austria (2024), KI-Umsetzungsplan 2024, [Translation], https://www.ki-strategie.at/umsetzungsplan-2024/ (accessed on 14 January 2025).
[47] FFG (2024), “Quantum Austria”, Austrian Research Promotion Agency, https://www.ffg.at/en/quantum-austria.
[17] France Digitale (2024), “From chips to apps, can Europe compete in generative AI?”, France Digitale, https://francedigitale.org/en/posts/report-generative-ai (accessed on 22 January 2025).
[13] French Government (2024), “IA: Notre ambition pour la France”, https://www.economie.gouv.fr/cge/commission-ia#:~:text=La%20Commission%20de%20l'intelligence%20artificielle%20a%20remis%20au%20pr%C3%A9sident,'intelligence%20artificielle%20(IA). (accessed on 22 January 2025).
[53] French Government (2022), “France 2030 : la stratégie pour l’électronique”, https://presse.economie.gouv.fr/12-07-2022-dossier-de-presse-electronique-2030/ (accessed on 18 January 2025).
[37] Gaia-X (2024), “About Gaia-X”, https://gaia-x.eu/about/ (accessed on 15 November 2024).
[45] GCS (2024), “Discover Gauss”, Gauss Centre for Supercomputing, https://www.gauss-centre.eu/about-us (accessed on 22 January 2025).
[24] German Federal Government (2023), “Driving progress with data”, https://bmdv.bund.de/SharedDocs/DE/Anlage/K/driving-progress-with-data.pdf?__blob=publicationFile (accessed on 25 January 2025).
[5] Government of Ireland (2024), Ireland’s National AI Strategy. AI – Here for Good – 2024 Refresh, Government of Ireland, https://enterprise.gov.ie/en/publications/publication-files/national-ai-strategy-refresh-2024.pdf.
[38] Government of Ireland (2021), Ireland’s National Recovery and Resilience Plan – Europe’s Contribution to Ireland’s Recovery, Government of Ireland, https://assets.gov.ie/162639/e5f1a2bf-35aa-4e25-9a8a-9b49c8b0b50f.pdf.
[27] Government of Ireland (2019), “Public service data strategy 2019-2023”, https://www.gov.ie/en/publication/1d6bc7-public-service-data-strategy-2019-2023/ (accessed on 22 January 2025).
[59] Government of Luxembourg (2025), Luxembourg’s AI Strategy, Ministry of Digitalisation, https://gouvernement.lu/en/publications/rapport-etude-analyse/minist-digitalisation/2025-luxembourg-ai-strategy-fr.html.
[55] Government of Poland (2023), Projekt uchwały Rady Ministrów w sprawie ustanowienia rządowego programu „Krajowe Ramy wspierania strategicznych inwestycji półprzewodnikowych“, [Translation], https://www.gov.pl/web/premier/projekt-uchwaly-rady-ministrow-w-sprawie-ustanowienia-rzadowego-programu-krajowe-ramy-wspierania-strategicznych-inwestycji-polprzewodnikowych (accessed on 18 March 2025).
[36] Government of Spain (2024), “Plan for the promotion of sectorial data spaces”, https://datos.gob.es/es/noticia/plan-de-impulso-de-los-espacios-de-datos-sectoriales (accessed on 15 January 2025).
[54] Government of Spain (2022), “PERTE de microelectrónica y semiconductores”, [English translation], https://planderecuperacion.gob.es/como-acceder-a-los-fondos/pertes/perte-de-microelectronica-y-semiconductores (accessed on 18 January 2025).
[35] HDN (2024), “Alle partijen in de financiële keten verbonden”, [Translation], HDN, https://hdn.nl/ (accessed on 6 November 2024).
[56] KIGEiT (2023), “Uchwała w sprawie „Krajowych Ram Wspierania Strategicznych Inwestycji Półprzewodnikowych””, [Translation], Polish National Chamber of Commerce for Electronics, https://kigeit.org.pl/2023/12/20/uchwala-w-sprawie-krajowych-ram-wspierania-strategicznych-inwestycji-polprzewodnikowych/ (accessed on 12 March 2025).
[28] KL (2022), “The municipalities’ data strategy”, https://www.kl.dk/oekonomi-og-administration/digitalisering-og-teknologi/politik-og-strategi/kommunernes-datastrategi (accessed on 18 January 2025).
[3] komunikace, O. (2024), “Czechia as a technological leader. Government approved the national strategy for artificial intelligence of the Czech Republic 2030”, https://www.mpo.gov.cz/en/guidepost/for-the-media/press-releases/czechia-as-a-technological-leader--government-approved-the-national-strategy-for-artificial-intelligence-of-the-czech-republic-2030--282278/ (accessed on 23 January 2025).
[57] Malta Enterprise (2024), “Driving innovation in Malta’s semiconductor industry”, https://maltaenterprise.com/ (accessed on 18 January 2025).
[14] McKinsey & Company (2024), “AI power: Expanding data center capacity to meet growing demand”, https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/ai-power-expanding-data-center-capacity-to-meet-growing-demand (accessed on 18 January 2025).
[26] Ministère de la transformation et de la fonction publiques (2021), “Cadre interministériel d’administration de la donnée”, Briefing, Ministère de la transformation et de la fonction publiques, Paris, https://www.numerique.gouv.fr/uploads/Cadre-interministeriel-donnee.pdf.
[60] Ministry for the Digital Transformation and the Public Function (2025), The Government launches the first Quantum Technologies Strategy of Spain with an investment of 800 million euros, https://digital.gob.es/en/comunicacion/notas-prensa/mtdfp/2025/04/2025-04-24 (accessed on 30 April 2025).
[44] Neistadt, M. (2025), “Briefing: AI factories”, European Parliament, Brussels, https://www.europarl.europa.eu/RegData/etudes/BRIE/2025/769492/EPRS_BRI(2025)769492_EN.pdf.
[49] Novo Nordisk Foundation (2024), “Denmark to build one of the world’s most powerful AI supercomputers, accelerating solutions to societal challenges”, https://novonordiskfonden.dk/en/news/denmark-to-build-one-of-the-worlds-most-powerful-ai-supercomputers-accelerating-solutions-to-societal-challenges/ (accessed on 25 January 2025).
[8] OECD (2024), OECD Artificial Intelligence Review of Germany, OECD Publishing, Paris, https://doi.org/10.1787/609808d6-en.
[10] OECD (2024), OECD Digital Economy Outlook 2024 (Volume 1): Embracing the Technology Frontier, OECD Publishing, Paris, https://doi.org/10.1787/a1689dc5-en.
[21] OECD (2022), “Data in an evolving technological landscape: The case of connected and automated vehicles”, OECD Digital Economy Papers, No. 346, OECD Publishing, Paris, https://doi.org/10.1787/ec7d2f6b-en.
[20] Politico (2025), “Trump’s $500B AI plan is ‘slap’ in the face for Europe”, Politico, https://www.politico.eu/article/us-500-ai-europe-donald-trump-global-leadership-eu-social-media-china/?utm_source=email&utm_medium=alert&utm_campaign=Trump%E2%80%99s%20%24500B%20AI%20plan%20is%20%E2%80%98slap%20in%20the%20face%E2%80%99%20for%20Europe (accessed on 22 January 2025).
[25] Portugal Digital (2022), “National data strategy”, https://portugaldigital.gov.pt/en/accelerating-digital-transition-in-portugal/get-to-know-the-digital-transition-strategies/national-data-strategy/ (accessed on 22 January 2025).
[50] Presidencia del Gobierno (2023), “El Consejo Asesor de Inteligencia Artificial aborda los avances en la Estrategia Nacional de IA financiados con el Plan de Recuperación”, [Translation], Presidencia del Gobierno, https://www.lamoncloa.gob.es/serviciosdeprensa/notasprensa/asuntos-economicos/Paginas/2023/060323-inteligencia-artificial.aspx#:~:text=El%20Consejo%20Asesor%20de%20Inteligencia%20Artificial%2C%20constituido%20en%20julio%20de,%C3%A9tico%20de%20la%20Intelig (accessed on 25 January 2025).
[51] Quantum Spain (2025), “The definitive boost to the quantum computing system in Spain”, https://quantumspain-project.es/en/quantum-spain/ (accessed on 12 March 2025).
[34] Smart Connected Supplier Network (2024), “About us”, https://smart-connected.nl/en (accessed on 6 November 2024).
[33] SolidLab (2024), “Welcome to SolidLab!”, https://solidlab.be/ (accessed on 3 December 2024).
[12] Ta, R. and N. Turner Lee (2023), “How language gaps constrain generative AI development”, https://www.brookings.edu/articles/how-language-gaps-constrain-generative-ai-development/ (accessed on 15 January 2025).
[16] Top 500 (2025), “Top 500. The List”, https://www.top500.org/ (accessed on 16 March 2025).
[29] Tourism Data Space (2024), “About the tourism data space”, https://www.tourism-dataspace.com/en/about-the-data-space (accessed on 15 January 2025).
[46] VSC (2024), “Homepage”, Vienna Scientific Cluster, https://vsc.ac.at/home/ (accessed on 5 November 2024).
← 1. Expected to be operating at full capacity by 2027. Notably, the project on technologies is being developed as part of the 2028 IPCEI for research and innovation in microelectronics.
← 2. For a comprehensive overview, please refer to: https://coe-dsc.nl/knowledge-base/community/data-sharing-initiatives-data-spaces/
← 3. The Quantum Technologies Strategy was published in April 2025 (Ministry for the Digital Transformation and the Public Function, 2025[60]).