Strategic Approach to Artificial Intelligence
Denmark · Strategy · The Ministry of Digital Affairs
Our summary
Denmark's government strategy for developing and using artificial intelligence responsibly across society. It applies to the public sector, businesses, and researchers, establishing three guiding principles centered on protecting citizens' fundamental rights, enabling competitive companies, and making Denmark a world leader in public sector AI use. The strategy funds four key initiatives: establishing a Digital Taskforce for public sector AI deployment, creating a research centre for responsible AI advice, developing secure Danish language models, and making Danish text data openly available.
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 Our notes summarise each one in plain language and do not quote the official text, which is the authority.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.
7 sections
Jump to a chapter
Publication cover and contents
The government commits to accelerating responsible AI use in Denmark through four new initiatives: establishing a Digital Taskforce to deploy AI solutions in the public sector, creating a research-based centre for advice on responsible AI use, developing secure transparent Danish language models, and making Danish data freely available. Denmark aims to be a global leader in applying AI within the public sector and maintains that AI development must always be citizen-centric and respect Danish values.
What is artificial intelligence and Danish vision
Denmark's AI strategy centres on three guiding principles: development must prioritise citizens' fundamental rights and Danish values, Danish companies must compete globally in responsible AI development and application, and Denmark must lead in public sector AI use to improve labour efficiency and administration. Four initiatives strengthen the foundation: the Digital Taskforce expands AI use across the public sector; a new centre provides research-based advice on responsible AI use; a secure platform will develop transparent Danish language models; and Danish text data will be released openly. Funding is allocated at DKK 62.5 million for initiatives 2, 3 and 4 in 2024-2027, DKK 40 million for initiatives 2 and 3 from the 2025 research reserve, and DKK 30.6 million for the Digital Taskforce in 2025-2027.
Digital Taskforce for Artificial Intelligence
Denmark's public authorities, working with local government bodies and regional assemblies, established a structured taskforce to coordinate AI development and expansion across the public sector, with a mandate to develop a vision with concrete objectives by 2030 for improving public services, reducing administrative burden, and freeing up workforce capacity, and to propose specific AI implementations for priority areas in government operations. The government will create a dedicated research and advisory facility to provide authorities and organisations with guidance on responsible AI deployment that aligns with democratic and ethical values, combining technical expertise with social and ethical perspectives and conducting independent research on generative AI, funded with approximately 20.7 million DKK from 2024-2027 plus 30 million DKK from the research reserve.
Secure platform for developing transparent Danish language models
A secure platform will be established by Danish Foundations Models, a collaboration between several Danish universities and the Alexandra Institute, for developing and training transparent Danish language models based on principles of transparency and data protection, providing open-source access and free access for Danish companies and authorities to develop and use specialised models, with DKK 20.7 million in funding for 2024-2027 plus DKK 10 million for research and innovation in 2025. Danish language text data from archives, libraries and parliamentary records will be standardised, made openly available, and hosted on a public platform to enable development of AI solutions that understand Danish language, culture and values, with approximately 21.1 million DKK funding from 2024-2027 plus an additional 30 million DKK annually for acquiring public data for training models.
AI adoption and public investment in Denmark
Denmark leads European corporate AI adoption at 15% in 2023 (compared to 8% EU average) and an estimated 28% in 2024, with over DKK 1 billion allocated across 18 ministerial areas for AI projects, particularly in research and education, and a Pioneer Centre for Artificial Intelligence that received DKK 354 million over 13 years for fundamental research and human-centric AI development. Denmark participates in European supercomputer collaboration and operates the Gefion supercomputer through public-private partnership, providing Danish researchers and companies with computing power for AI research and development in three priority areas: pharmaceutical and biotechnology research, green energy transition, and quantum computing development.
Public attitudes toward AI
A February 2024 public consultation of 356 representative Danes showed majority support for AI's benefits to companies (78%), society (61%), and individuals (55%) over 20 years, with participation correlating to increased support and moderation of views, though most participants also believed AI regulation necessary. Denmark's AI approach is based on credibility and human-centric development to prevent realizing technology benefits at the expense of fundamental rights, safety or health. The EU's Artificial Intelligence Act, adopted in June 2024 and in force from 1 August 2024, establishes binding risk-based rules where higher-risk systems face more restrictive requirements, aiming to improve internal market functioning, promote human-centric AI adoption, and protect citizens from inappropriate and harmful uses.
Risk-based implementation of the AI Act
The AI Act applies a risk-based framework where unacceptable-risk systems are prohibited (effective February 2025), high-risk systems face strict requirements including human oversight (phased in August 2026 to August 2027), limited-risk systems must be transparent, and minimal-risk systems face no obligations; all developers, deployers, importers, distributors and producers must comply with Denmark's Agency for Digitisation overseeing national compliance and providing guidance. The Agency leads implementation by establishing governance structures, publishing information and guidance at digst.dk with regular stakeholder involvement, creating instructional materials on responsible generative AI use for business, authorities and citizens, and publishing periodic reviews of AI experience from signature projects. The Danish Data Protection Agency has issued separate guidance for public authorities developing or using AI with personal data processing.
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 8 of 35
Choose a theme to see only its controls. About the themes
Risk and scope 2 of 4
- Risk management system, not covered
- Impact assessment, not covered
Data and privacy 0 of 4
- Data governance and quality, not covered
- 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 1 of 4
- Accuracy, robustness and cybersecurity, not covered
- Bias and non-discrimination, not covered
- Testing and evaluation, not covered
Assurance and lifecycle 0 of 4
- Conformity assessment and certification, not covered
- Registration and notification, not covered
- Incident and breach reporting, not covered
- Post-market monitoring and corrective action, 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 3 of 6
- Supply chain and provider-deployer duties, not covered
- Training and AI literacy, not covered
- Innovation support and sandboxes, not covered
Enforcement and sectors 1 of 5
- 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.
Digital Taskforce on Artificial Intelligence 4 controls
- ObligationApplies to: Public authorityDrafted by the Index
The government has established a taskforce to develop and deploy AI solutions across the public sector.
Digital Taskforce on Artificial IntelligenceOfficial text of Digital Taskforce on Artificial Intelligence (opens in a new tab) - ObligationApplies to: Public authorityDeadline Drafted by the Index
The taskforce must create a vision with targets for freeing up staff time by 2030.
Digital Taskforce on Artificial IntelligenceOfficial text of Digital Taskforce on Artificial Intelligence (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
The taskforce must identify implementation priorities and propose AI solutions for public sector adoption.
Digital Taskforce on Artificial IntelligenceOfficial text of Digital Taskforce on Artificial Intelligence (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
The taskforce must remove organisational and legal barriers to AI adoption in the public sector.
Digital Taskforce on Artificial IntelligenceOfficial text of Digital Taskforce on Artificial Intelligence (opens in a new tab)
Secure platform for Danish language models 3 controls
- ObligationApplies to: Public authorityDrafted by the Index
A secure platform for developing transparent Danish language models must be created.
Secure platform for Danish language modelsOfficial text of Secure platform for Danish language models (opens in a new tab) - Applies to: AnyoneDrafted by the Index
Danish businesses and authorities should be able to access and develop smaller specialised models on the platform.
Secure platform for Danish language modelsOfficial text of Secure platform for Danish language models (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
The platform will distribute models as open source to support Danish service delivery and business competitiveness.
Secure platform for Danish language modelsOfficial text of Secure platform for Danish language models (opens in a new tab)
EU AI Act implementation and risk-based approach 4 controls
- ProhibitionApplies to: ProviderDeadline was Drafted by the Index
Providers must prohibit AI systems identified as posing unacceptable risk to health, safety or rights, with bans taking effect from February 2025.
EU AI Act implementation and risk-based approachOfficial text of EU AI Act implementation and risk-based approach (opens in a new tab) - ObligationApplies to: ProviderDeadline was Drafted by the Index
Providers of high-risk systems must implement strict safeguards including human oversight, taking effect from August 2026.
EU AI Act implementation and risk-based approachOfficial text of EU AI Act implementation and risk-based approach (opens in a new tab) - ObligationApplies to: ProviderDrafted by the Index
Providers of lower-risk systems must meet transparency or minimal requirements as scaled according to risk level.
EU AI Act implementation and risk-based approachOfficial text of EU AI Act implementation and risk-based approach (opens in a new tab) - ObligationApplies to: Public authorityDrafted by the Index
Denmark's Agency for Digitisation shall serve as the national coordinating authority for oversight and guidance on AI regulation.
EU AI Act implementation and risk-based approachOfficial text of EU AI Act implementation and risk-based approach (opens in a new tab)
Denmark and AI in an International Perspective 1 control
- ObligationApplies to: Public authorityDrafted by the Index
Denmark should contribute to European regulatory frameworks and international governance forums to promote trustworthy, human-centered and environmentally responsible AI deployment.
Denmark and AI in an International PerspectiveOfficial text of Denmark and AI in an International Perspective (opens in a new tab)
Change history
No changes recorded since we began monitoring this policy.