The European Union is building its AI policy on three connected layers: the AI Continent Action Plan (2025) to build capacity, the Tech Sovereignty Package (2026) to secure critical infrastructure, and regulation led by the AI Act (2024) and simplified by the Digital Omnibus. Click any item for more detail, or read the full text below.
building capacity
AI Continent Action Plan (2025)
The European Commission's plan to build AI compute capacity, widen access to high-quality data, and speed up AI adoption and skills across the single market. Click any item above for details.
strengthening infrastructure
Tech Sovereignty Package (2026)
A package of initiatives aimed at strengthening EU control over critical digital infrastructure — from cloud and chips to energy and open-source AI.
ensure AI safety
AI Act (2024)
The EU's horizontal regulation for artificial intelligence, setting risk-based obligations for providers and deployers of AI systems.
Digital Omnibus (2026)
A simplification package that streamlines overlapping digital-rulebook obligations — AI Act compliance included — for businesses operating in the EU.
Data Union Strategy (2025)
an action plan designed to unlock high-quality data for artificial intelligence (AI), simplify existing digital regulations, and protect Europe's data sovereignty. Building upon the 2020 European Strategy for Data, it shifts the EU's focus from creating foundational rules to driving practical, large-scale implementation across industries. It does so by Scaling up access to Data for AI (Common European Data Spaces and Data Labs), Streamlining Data Rules and Safeguarding Data Sovereignty.
Apply AI Strategy (2025)
The deployment counterpart to the Data Union Strategy focuses on real-world AI adoption across European industries and the public sector, promoting an "AI-first" culture and a "buy European" approach—particularly for open-source AI solutions.
related:
protect personal data
GDPR (2018)
The General Data Protection Regulation (GDPR) is a comprehensive privacy and data protection law enacted by the European Union (EU) that took effect on May 25, 2018. It sets strict rules for how organizations collect, store, process, and share the personal data of individuals in the EU and European Economic Area (EEA), while giving individuals enforceable control over their information. Since AI heavily relies on data, GDPR is a foundational building block for AI policy.
DSA(2022)
The Digital Services Act (DSA) governs online intermediaries and platforms—such as social media networks, online marketplaces, app stores, and search engines. While GDPR protects personal data and privacy, the DSA regulates online content, platform accountability, user safety, and algorithmic transparency. AI chatbots and other AI applications are subject to the DSA.
EU AI policy in depth
Last updated: 21 November 2026. By Lambert Hogenhout.
AI Continent Action Plan (2025)
The European Commission's plan to build AI compute capacity, widen access to high-quality data, and speed up AI adoption and skills across the single market. It has five pillars:
AI Factories and Gigafactories. Builds Europe's AI compute capacity: 19 AI Factories and 13 Antennas are being set up across the EU, backed by at least 9 new AI-optimised EuroHPC supercomputers that more than triple EuroHPC's AI capacity, for roughly €10 billion in funding through 2027. Up to 7 AI Gigafactories, about four times more powerful and each housing over 100,000 advanced AI processors, are being procured via the InvestAI initiative to train frontier, trillion-parameter models, with up to €10 billion in EU/national funding expected to unlock at least €20 billion in private investment.
High-quality data. Delivered through the Data Union Strategy: makes high-quality data easier for businesses, researchers and public bodies to access, share and reuse for training AI, while simplifying the EU's patchwork of data rules. Data Labs inside the AI Factories curate sector-specific datasets in healthcare, mobility, manufacturing and more.
Accelerating AI uptake. Moves AI out of pilots and into everyday use across the economy and public administration, with practical support, testing infrastructure and funding, especially for SMEs. Carried out mainly through the Apply AI Strategy.
Skills and talent. International partnerships to recruit AI talent, fellowships for top AI students, researchers and professionals, and a planned EU-wide AI Skills Academy, plus a pilot generative-AI university degree and reskilling through the European Digital Innovation Hubs (EDIHs).
Support for implementing the AI Act. The AI Act Service Desk, launched October 2025, gives companies free, tailored guidance on their obligations. A related AI omnibus simplification package, proposed November 2025 and adopted June 2026 (in force since 27 July 2026), made targeted amendments to ease some of the Act's administrative burden without reopening its core rules.
Tech Sovereignty Package (2026)
A package of initiatives aimed at strengthening EU control over critical digital infrastructure, from cloud and chips to energy and open-source AI.
CADA (Cloud and AI Development Act). A proposed regulation to strengthen Europe's sovereignty and competitiveness in cloud and AI, targeting at least a tripling of EU data-centre capacity within 5-7 years. It funds frontier, industrial and physical AI research through grand challenges; speeds up data-centre permitting and access to energy, land, water and financing; and creates an EU-wide cloud/AI sovereignty framework with four assurance levels, from data stored in the EU up to full supply-chain transparency, alongside a common EU procurement framework.
Chips 2.0. The next phase of the EU Chips Act, aiming to strengthen Europe's semiconductor industry, reduce strategic dependencies and boost domestic demand. The original Chips Act helped mobilise over €52 billion in public and private investment and an estimated 46,000 jobs; the global chip market is projected to reach €1.37 trillion by 2030, with AI-related components driving about 70% of growth. Chips 2.0 proposes permitting decisions within 12 months, grand challenges such as AI chips, strategic partnerships, demand accelerators, state aid for first-of-a-kind projects, a Semiconductor Regions of Excellence label, and a B2B supply-chain platform.
Energy. The Strategic Roadmap for Digitalisation and AI in Energy (COM(2026) 501) aims for a fully digitalised, AI-enabled EU energy system by 2035, built on three pillars: energy for AI (data-centre power demand projected to rise from roughly 12 GW in 2025 to about 28 GW by 2030), digitalisation and AI for energy, and data for AI.
Open Source. The EU Open Source Strategy places open source at the centre of technological sovereignty, promoting European open alternatives across operating systems, cloud, cybersecurity, semiconductors and AI. It proposes scaling the Open Internet Stack catalogue, an Open Source Maintenance Instrument, mapping of critical dependencies, procurement guidance, and a stronger Commission Open Source Programme Office.
AI Act (2024)
The EU's horizontal regulation for artificial intelligence, setting risk-based obligations for providers and deployers of AI systems.
Digital Omnibus (2026)
A simplification package that streamlines overlapping digital-rulebook obligations, AI Act compliance included, for businesses operating in the EU.
Data Union Strategy (2025)
An action plan designed to unlock high-quality data for AI, simplify existing digital regulations, and protect Europe's data sovereignty. Building on the 2020 European Strategy for Data, it shifts the EU's focus from creating foundational rules to practical, large-scale implementation, by scaling up access to data for AI (Common European Data Spaces and Data Labs), streamlining data rules and safeguarding data sovereignty.
Apply AI Strategy (2025)
The deployment counterpart to the Data Union Strategy. It focuses on real-world AI adoption across European industries and the public sector, promoting an "AI-first" culture and a "buy European" approach, particularly for open-source AI solutions.
Related legislation: GDPR (2018) and DSA (2022)
The General Data Protection Regulation (GDPR), in effect since 25 May 2018, sets strict rules for how organizations collect, store, process and share the personal data of individuals in the EU and EEA. Since AI relies heavily on data, GDPR is a foundational building block for AI policy.
The Digital Services Act (DSA) governs online intermediaries and platforms such as social networks, marketplaces, app stores and search engines. While GDPR protects personal data, the DSA regulates online content, platform accountability, user safety and algorithmic transparency. AI chatbots and other AI applications are subject to the DSA.
Frequently asked questions
What is the AI Continent Action Plan?
The European Commission's 2025 plan to build AI compute capacity, widen access to data, accelerate AI uptake, develop skills and support AI Act implementation.
What is the EU Tech Sovereignty Package?
A 2026 package covering the Cloud and AI Development Act (CADA), Chips 2.0, the digitalisation and AI in energy roadmap, and the EU Open Source Strategy.
What is the EU AI Act?
The EU's 2024 risk-based regulation for AI systems, with obligations for providers and deployers.
What is the Digital Omnibus?
A simplification package that streamlines overlapping digital-rulebook obligations, including AI Act compliance.
How do GDPR and the DSA relate to AI?
GDPR governs the personal data AI depends on; the DSA governs online platforms and the AI applications offered through them.