Norway - National AI Strategy

Overview

The National Strategy for Artificial Intelligence (published 14 January 2020) sets out Norway's ambition to harness AI for public value, industry competitiveness and social welfare while safeguarding trust, human rights and privacy. The strategy is available in English and Norwegian; the English version is published on the Government of Norway website and the full text is downloadable as a PDF. The Government frames AI as both an economic opportunity and a governance challenge: it seeks to ensure that Norway develops world-class AI infrastructure, good language resources, adequate computing power and robust networks. At the same time the strategy emphasises that AI must be developed and applied responsibly, with Norway leveraging its high level of public trust and digital maturity to lead in 'human-friendly' AI. Key actions include facilitating access to data, supporting research and skills, piloting regulatory sandboxes, and coordinating oversight by relevant supervisory authorities. See the official strategy document: The National Strategy for Artificial Intelligence (Government of Norway) and the full PDF: English PDF: National Strategy for Artificial Intelligence (2020).

Definitions

The strategy provides working definitions and explanatory material for artificial intelligence and related concepts. It describes AI broadly to include data-driven machine learning and algorithmic systems used to perform tasks that would normally require human intelligence (pattern recognition, prediction, classification, decision support). The document clarifies distinctions between models, data, systems and services, and explains relevant technical concepts at a level intended for policymakers and public agencies. Definitions emphasise that governance must address systems (not only code): datasets, training methods, lifecycle deployment, monitoring and human oversight are all treated as part of the AI system. The strategy aligns with common EEA/EU terminology to facilitate international cooperation and regulatory interoperability.

Governance and Institutional Framework

The strategy designates the ministry responsible for digitalisation and public governance as the lead coordinating authority and calls for strengthened inter‑ministerial cooperation. It asks supervisory authorities (for example, the Norwegian Data Protection Authority (Datatilsynet)) to oversee AI systems within their sectoral remit and to provide guidance on privacy, fairness, and transparency. The document proposes the establishment of an advisory body and regulatory sandboxes — including a privacy-focused sandbox — to test novel AI applications under supervised conditions. The strategy stresses the role of public procurement, government research councils and university institutions in delivering centres of excellence and shared services (language datasets, HPC access). Coordination between national agencies is to be complemented by proactive engagement at EU and OECD levels to align standards and enforcement practices.

Key Focus Areas

The strategy organizes policy action around several mutually reinforcing focus areas:

• Data and data management: improving access to high-quality, interoperable datasets (open public data, registries, and language resources for Norwegian and Sami languages) while protecting privacy and confidentiality.
• Regulations and regulatory sandboxes: reviewing and refining rules that unintentionally hinder digital innovation, and piloting sandboxes in domains such as autonomous mobility and data-protection-sensitive AI.
• Infrastructure: securing high-speed networks (5G), national and international high-performance computing resources, and resilient data centre capacity.
• Research & higher education: boosting funding for basic and applied AI research and embedding AI topics across education to build interdisciplinary expertise.
• Skills & workforce: developing retraining, upskilling, and flexible further education programmes to manage labour market transitions.
• Innovation and public sector adoption: using procurement and public-private collaboration to stimulate AI development in priority sectors (health, maritime, energy, public services).
• Trustworthy AI: articulating and promoting ethical principles (transparency, human oversight, privacy protection, cautious testing, explainability) and insisting supervisory oversight in deployment.

The strategy treats these areas as integrated: data and infrastructure enable research and industry, while ethics and regulation enable trustworthy adoption.

Implementation Framework

The strategy outlines implementation mechanisms rather than binding legal obligations. Implementation levers include budgetary support for research (through the Research Council of Norway and grant programs), targeted competence-building initiatives (education curricula, short courses and workplace training), creation of shared public datasets and language resources, and investments in HPC and network infrastructure. The strategy also recommends clarifying sectoral regulation that intersects with AI (healthcare, transport, energy) and encourages the use of regulatory sandboxes where legal uncertainty or innovation risk justifies supervised experimentation. Ministries are asked to identify priority measures, oversee timetable and funding, and coordinate with supervisory agencies to ensure legal compliance. A central theme is that existing laws (e.g., the Personal Data Act/GDPR, Public Administration Act) remain applicable and supervisory authorities should exercise their oversight powers in AI matters.

Monitoring and Evaluation

Monitoring is framed around periodic assessment of progress against objectives (research capacity, infrastructure access, uptake across sectors, ethical adoption). The strategy envisages public reporting and an iterative approach: outcomes from sandboxes and advisory bodies should feed back into policy and regulatory clarification. It also calls on national audit and supervisory institutions to evaluate whether agencies and regulated entities implement the strategy’s principles in practice. Indicators suggested include the number of research projects, HPC capacity made available, uptake metrics in priority sectors, and qualitative assessments of transparency and ethical compliance in public deployments. The aim is to allow course-correction while maintaining oversight of rights and safety.

Penalties, Liability, and Appeals

The strategy itself is a policy/strategic document and does not create new penalties. Instead, it affirms that existing legal frameworks (notably the Personal Data Act and sectoral statutes) apply to AI systems and that supervisory authorities retain their enforcement powers under those laws. Where enforcement or liability issues arise in AI deployments, the strategy expects the relevant supervisory authority (for example, Datatilsynet for privacy matters) and sectoral regulators to use established enforcement tools, including investigations, orders, and administrative fines where permitted by law. The strategy also highlights the need for clear responsibility chains in procurement and deployment so that liability and redress can be established under existing civil or administrative law.

Relationship to Other Instruments

The strategy explicitly situates itself alongside existing Norwegian laws and international instruments. It references the Personal Data Act (implementing the GDPR), the Public Administration Act, the Archival Act, sectoral health and transport regulation, and international commitments (EEA cooperation). It anticipates the need to align national practice with the EU's AI regulatory framework and related OECD and Council of Europe instruments. The document promotes regulatory sandboxes as a way to reconcile innovation and legal compliance and advises ministries to review domestic rules that could unintentionally block beneficial AI uses while preserving fundamental rights.

International Alignment

Norway commits to active international cooperation. The strategy states that Norway will participate in EU, OECD and Council of Europe processes to influence standards and ensure interoperability. It highlights Norway's EEA relationship and the practical need to follow developments such as the EU's Artificial Intelligence Act and related guidance. The strategy also encourages collaboration on research and infrastructure with international partners (HPC, Horizon programmes) and notes that international alignment is critical for market access, procurement and cross-border data flows.

Implementation Timeline

Phase	Actions	Indicative Timing
Immediate (2020)	Publish strategy, launch advisory body, begin regulatory review and open data initiatives, establish sandboxes.	2020–2021
Short term	Fund research centres, launch skills initiatives and language resource projects, secure HPC partnerships.	2020–2022
Medium term	Scale public procurement use of AI, evaluate sandboxes, strengthen supervisory guidance and cross‑agency coordination.	2022–2024
Ongoing	Monitor uptake, iterate policy, participate in EU/EEA rule-making and align domestic practice.	2020 onward

Compliance Checklist

Checklist Item	Who	Notes
Assess data protection impact (DPIA)	Deployers/Controllers	Follow Datatilsynet guidance where personal data involved.
Ensure transparency and human oversight	Providers/Deployers	Document decision-making chains and provide user information.
Use privacy‑by‑design	Developers	Embed data minimisation and security early in design.
Engage with supervisory sandboxes	Innovators	Apply to Datatilsynet or sector sandboxes for regulated testing.
Record-keeping and documentation	Providers	Maintain technical documentation to support audits.

Sources and References

Source	Type
The National Strategy for Artificial Intelligence (Government of Norway – English page)	Primary Source
English PDF: National Strategy for Artificial Intelligence (2020)	Primary Source
Norwegian Data Protection Authority (Datatilsynet) – English	Primary Source (supervisory guidance)
European Commission – AI Act / EU developments	Context / International Alignment