European approach to artificial intelligence
European Union · Framework · European Commission
Our summary
The EU's comprehensive policy framework for artificial intelligence, launched in 2018, aims to promote both technological excellence and trustworthiness through regulation, investment, and capacity-building. It encompasses multiple initiatives including the AI Act (a binding legal framework entered into force in 2024), the AI Continent Action Plan for large-scale infrastructure development, and the Apply AI Strategy to increase adoption across industries and public services. The approach balances innovation promotion with safeguards for safety, fundamental rights, and legal certainty.
Written by the Index in our own words. It is not legal advice; the official text is the authority.
Inside this policy
Every section of this policy in order. Our notes summarise each one in plain language and do not quote the official text, which is the authority.
Drafted by the Index An automated summary of the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
8 sections
Jump to a chapter
1 Introduction, embracing change
Artificial intelligence is defined as systems that analyse their environment and take autonomous actions to achieve specific goals, transforming society across sectors like healthcare, agriculture, and manufacturing. The EU aims to strengthen technological capacity and AI adoption through research investment, support social and economic transition via education and labour market adaptation, and establish an ethical and legal framework aligned with EU values and fundamental rights. The Commission intends to coordinate with Member States on a joint AI plan to align investments and establish best practices.
2 The EU's position in a compet itive international landscape
The United States invested approximately 970 million euros in unclassified AI research in 2016 and China targets global leadership by 2030, while private sector AI investment in 2016 was 2.4 to 3.2 billion euros in Europe, 6.5 to 9.7 billion euros in Asia and 12.1 to 18.6 billion euros in North America. Europe has world-class AI research capacity, innovative entrepreneurs and leading robotics and manufacturing sectors, but adoption of digital and AI technologies among European companies remains low. The EU invested approximately 700 million euros in research and development frameworks from 2014 to 2020, with 2.1 billion euros from private robotics partnerships and roughly 1.1 billion euros for AI-related research and innovation under Horizon 2020 during 2014 to 2017, while additionally launching initiatives on efficient electronic components and systems, high-performance computing, quantum technologies and brain mapping research.
3 The way forward: an EU initiative on AI
4 sections
Outlines the European Union's strategic framework on artificial intelligence, covering four main areas: building technological and industrial capacity, managing socioeconomic transitions, establishing ethical and legal safeguards, and coordinating efforts across member states and stakeholders.
3.1 Boosting the EU's technological and industrial capacity and AI uptake across the economy
7 billion for 2018-2020 to support approximately 1,000 breakthrough projects and 3,000 feasibility awards through the European Innovation Council pilot, strengthen AI-focused research centres, and support Digital Innovation Hubs and an AI-on-demand platform providing access to AI resources. 5 billion to research, excellence centres and uptake initiatives. After 2020, it proposes to upgrade excellence centres, fund explainable AI and machine learning research, establish additional Digital Innovation Hubs with testing facilities and regulatory sandboxes, support AI adoption through co-investment with Member States, create a data sharing support centre, and support enabling technologies.
3.2 Preparing for socioeconomic changes
New technologies including AI are transforming labour markets, requiring the EU and Member States to develop basic digital and complementary skills for all Europeans, support workers in roles at risk of automation through retraining and skill acquisition opportunities funded by European Structural and Investment Funds and private investment, and train more AI specialists while promoting diversity and attracting international talent. EU actions include establishing targeted retraining programmes for roles at risk of automation, commissioning analysis of labour market changes and skills gaps, supporting digital traineeships, fostering business-education partnerships, encouraging social partners to address AI's employment impacts, incorporating AI across educational curricula, and expanding an EU fund for workers affected by digitisation and automation.
3.3 Ensuring an appropriate ethical and legal framework
The EU's regulatory framework based on fundamental rights and GDPR protections, including the right not to be subject to automated decision-making alone in certain situations, provides the foundation for sustainable AI development. The Commission will develop draft AI ethics guidelines by year-end addressing work, fairness, safety, security, inclusion, algorithmic transparency, and impacts on fundamental rights, invite stakeholder contributions, assess adequacy of safety and civil liability frameworks for AI-enabled products, and may address gaps through safety standards development and review of product liability and machinery directives. Consumers in AI-enabled transactions must receive clear information on features and properties, control their data, know whether interacting with machines or humans, and be informed how to reach humans and verify or correct automated decisions; the Commission will establish a European AI Alliance by year-end, issue product liability guidance by mid-2019, publish a liability and safety framework report by mid-2019, and support explainable AI research and stakeholder capacity-building.
3.4 Joining forces
The Commission calls on all Member States to develop AI strategies with investment plans, notes that 24 Member States and Norway have committed to collaborate on a joint plan, and plans to establish a broad multi-stakeholder European AI Alliance by July 2018 involving businesses, consumer organisations, trade unions and civil society to develop and share AI practices and monitor AI uptake and technical capabilities. Recognising that global solutions are necessary as AI crosses borders easily, the Commission commits to working with Member States by year-end on coordinated investment impact, government preparation for AI transformation, legal and ethical matters, while systematically monitoring AI developments, uptake, labour market impacts and capabilities.
4 Conclusion
The EU must leverage its scientific and industrial foundations to lead the AI revolution while keeping fundamental rights and human-centred design central, securing adequate resources through mechanisms including the Commission's forum for strategic value chain funding and cross-regional partnerships supported by the Horizon 2020 work programme. AI systems should enable human understanding of their decision-making to enhance transparency and reduce bias, and the EU is addressing legal and ethical frameworks including intellectual property questions, liability through the Product Liability Directive, and standardization needs for interoperability and fair competition through international and Member State cooperation.
Controls
What this policy requires, forbids, grants or penalises, one control at a time and in our own words. Each control links to the official text, which is the authority. This is not legal advice.
Drafted by the Index Drafted automatically from the official text, not yet reviewed by a person.Reviewed Checked by a person against the official text.
Themes covered 5 of 35
Choose a theme to see only its controls. About the themes
Risk and scope 0 of 4
- Risk classification, not covered
- Prohibited practices, not covered
- Risk management system, not covered
- Impact assessment, not covered
Data and privacy 1 of 4
- Data subject rights and consent, not covered
- Cross-border data transfer, not covered
- Copyright and training data, not covered
Transparency and records 1 of 4
- Labelling of synthetic content, not covered
- Technical documentation, not covered
- Record keeping and logging, not covered
Technical safeguards 0 of 4
- Human oversight, not covered
- Accuracy, robustness and cybersecurity, not covered
- Bias and non-discrimination, not covered
- Testing and evaluation, not covered
Assurance and lifecycle 1 of 4
- Conformity assessment and certification, not covered
- Registration and notification, not covered
- Incident and breach reporting, not covered
Rights and protected groups 0 of 4
- Automated decision-making rights, not covered
- Children and vulnerable groups, not covered
- Biometric identification and surveillance, not covered
- Whistleblowing, complaints and redress, not covered
Roles and organisation 2 of 6
- Supply chain and provider-deployer duties, not covered
- General purpose and foundation models, not covered
- Public sector use and procurement, not covered
- Innovation support and sandboxes, not covered
Enforcement and sectors 0 of 5
- Enforcement, penalties and regulators, not covered
- Sector rules: health, not covered
- Sector rules: finance and insurance, not covered
- Sector rules: employment, not covered
- Sector rules: education, not covered
The full list of controls could not be loaded, so only the first 12 are shown.
3.1 3 controls
- ObligationApplies to: AnyoneDrafted by the Index
Public and private sectors should jointly increase AI investment across the EU to at least 20 billion euros by 2020 and over 20 billion annually thereafter.
- ObligationApplies to: Public authorityDeadline was Drafted by the Index
The Commission should contribute 1.5 billion euros to AI research and innovation under Horizon 2020 by 2020.
- ObligationApplies to: Public authorityDrafted by the Index
The Commission should work with Member States to coordinate and accelerate AI investment plans.
3.2 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
Member States bear responsibility for helping all Europeans develop basic digital skills and complementary abilities that machines cannot replace, such as critical thinking and creativity.
- ObligationApplies to: Public authorityDrafted by the Index
The EU should support workers in roles likely to be transformed or eliminated by automation, robotics and AI to remain in the labour market.
- ObligationApplies to: Public authorityDrafted by the Index
The EU should train more AI specialists and create conditions to retain them and attract international talent.
3.3 2 controls
- ObligationApplies to: ProviderDrafted by the Index
AI systems should be developed to be transparent and explainable so that people can understand how they work.
- ObligationApplies to: Public authorityDrafted by the Index
The Commission should closely monitor how the General Data Protection Regulation applies to AI and call on data protection authorities to do likewise.
3.4 4 controls
- ObligationApplies to: Public authorityDrafted by the Index
Every Member State should develop an AI strategy that includes an investment plan.
- ObligationApplies to: Public authorityDrafted by the Index
The Commission should facilitate dialogue among Member States, coordinate AI action plans, and work to prevent fragmentation of the single market.
- ObligationApplies to: Public authorityDrafted by the Index
The Commission should establish and support a broad multi-stakeholder European AI Alliance to share best practices and encourage investment in AI development.
- ObligationApplies to: Public authorityDrafted by the Index
The Commission should monitor AI application uptake across the economy, identify shifts in industrial value chains, and track societal and legal developments.
Change history
- Official text wordingSource for this change (opens in a new tab)
- Official text wordingSource for this change (opens in a new tab)
- Official text wordingSource for this change (opens in a new tab)