No future for Nordic healthcare without AI
A position paper by The Nordic University Hospital Alliance (NUHA)
____________________
Artificial intelligence must become a core capability of healthcare. Faced with ageing populations, growing numbers of people living with chronic diseases, increasing demand for advanced treatments and persistent workforce shortages, healthcare systems across Europe need new ways to improve outcomes and use resources wisely. AI can help, but only if it is developed, validated and implemented in ways that benefit patients, support healthcare professionals and improve the quality of care.
The question is no longer whether AI will become part of healthcare. The question is who will shape its development and ensure that it reflects European values of trust, equity and public responsibility.
If Europe wants trusted AI in healthcare, university hospitals must play a leading role. And the Nordic region offers a unique opportunity to show how this can be done.
A unique Nordic opportunity
AI depends on more than sophisticated algorithms. It depends on access to high-quality data, secure and trusted infrastructures, clinical expertise and robust methods for evaluation and validation.
The Nordic countries already hold a strong position. We have outstanding public healthcare systems, high-quality health data, trusted public institutions and a long tradition of collaboration between clinical care, research and society. Few regions combine these advantages to the same extent.
At the same time, Europe is creating the framework for a new era of healthcare innovation through initiatives such as the European Health Data Space (EHDS) and the AI Act. These initiatives can help unlock the full potential of health data while safeguarding privacy, accountability and public trust.
The challenge now is implementation. How do we translate European ambitions into better healthcare for patients?
The role university hospitals play
University hospitals occupy a unique position at the intersection of patient care, education, research and innovation. They provide environment where new knowledge is generated, where technologies can be tested under real-world clinical conditions and where future healthcare professionals are educated and trained.
This makes university hospitals uniquely suited to lead the development, validation and implementation of AI. This leadership does not necessarily mean that university hospitals should own or manage the underlying data or digital infrastructure. Rather, their role is to bring clinical expertise, validation capacity and a shared healthcare perspective to collaborative Nordic structures.
University hospitals also play a pivotal role in creating data-driven healthcare systems, where knowledge generated through clinical care can continuously improve research, quality of health services and future therapeutic interventions. While AI may be the most visible innovation, its success depends on the effectiveness contingent upon the underlying systems that support this technology: reliable data, common standards, secure information sharing and rigorous clinical validation.
Trusted AI is therefore not only a technological challenge. It is an organisational and clinical challenge requiring collaboration across institutions, regions and countries.
From regulation to real-world impact
Europe has made significant progress in establishing regulatory frameworks for trustworthy AI. The next step is to ensure that these frameworks deliver tangible benefits for patients and healthcare professionals.
Healthcare systems do not need more isolated pilot projects. They need to implement AI at scale through shared environments where solutions can be tested, evaluated and continuously improved across different healthcare settings and patient populations.
By working together, Nordic university hospitals can build such shared test and learning environment. As a network, they offer clinical expertise, patient diversity and operational scale required to assess whether AI solutions are safe, effective and valuable before they are broadly adopted.
This would not only benefit the Nordic countries. It would provide a practical model for implementing trusted AI across Europe.
Three priorities for trusted AI in healthcare
1. Build a stronger Nordic health data foundation
Effective AI requires access to high-quality, reliable and interoperable health data. The value of AI ultimately depends on the data that underpins it. NUHA can play a role in bringing together and strengthening access to relevant health data across the Nordic region, while respecting national responsibilities, legal frameworks and patient privacy.
Nordic countries should strengthen collaboration around key datasets, clinical registries and secure cross-border data sharing. Information should be documented in ways that first and foremost provide support to patient care, while also enabling research, quality improvement and AI-enabled learning when appropriate.
A stronger data foundation is not the ultimate objective. It is the prerequisite for trustworthy AI.
2. Create Nordic infrastructures for AI evaluation
Healthcare should not adopt AI because it is new.
AI should be implemented because it has been demonstrated to improve patient outcomes, support healthcare professionals and enhance healthcare quality. Achieving this requires systematic testing and real-world evaluation.
Nordic university hospitals can provide shared infrastructures where AI solutions are assessed across different healthcare systems and patient populations before being scaled. Such infrastructures can be built and governed jointly with relevant national and Nordic data and digital health organisations, reflecting the different legal and organisational frameworks across the Nordic countries. This creates stronger evidence, greater transparency and higher levels of trust.
A practical example is the Nordic AI Health Map, which brings together AI initiatives and projects from across the Nordic countries on a shared platform. By making existing work more visible and easier to navigate, such an infrastructure can help healthcare organisations identify relevant initiatives, share experience and build on existing evidence rather than developing similar solutions in isolation.
3. Establish a common framework for trusted AI
Trust cannot be achieved through regulation alone.
Safe patient care, transparency, accountability, clinical oversight and continuous monitoring must be embedded throughout the lifecycle of AI solutions. Common approaches to governance, evaluation and validation are essential if healthcare organizations and patients are to trust the technologies they use. This also requires greater alignment of governance and data protection frameworks across the Nordic region. Where possible, shared approaches to legislation, GDPR implementation and the secondary use of health data should make it easier to collaborate across borders while maintaining strong safeguards for patients and citizens.
A shared Nordic framework could help ensure that AI solutions are not only innovative, but also reliable, safe and aligned with the values of publicly funded healthcare.
A question of healthcare sovereignty
The global race for healthcare AI is accelerating. Much of the investment and technological development is taking place outside Europe.
International collaboration and partnerships with industry are essential. But Europe must also retain the ability to shape how AI is developed and used in clinical practice.
If healthcare systems become entirely dependent on technologies developed elsewhere, they risk losing influence over decisions that directly affect patients, clinical workforce and healthcare delivery.
The Nordic countries have assets that Europe needs: trusted healthcare institutions, advanced health data ecosystems and a long tradition of public trust. Together, these strengths can help Europe build AI that is both globally competitive and aligned with European values.
An invitation
NUHA does not seek to create another discussion forum.
Our ambition is practical: to create a Nordic platform where healthcare providers, researchers, policymakers, patients and industry partners can work together to develop, validate and scale trustworthy AI. This is not a task for a handful of hospitals alone – it belongs to the whole Nordic healthcare community.
The Nordic countries have a long history of solving healthcare challenges through collaboration. AI offers an opportunity to build on that legacy, because no Nordic country can achieve the necessary scale alone.
The question is no longer whether AI belongs in healthcare.
The question is whether Europe – together - will shape the future of AI-supported healthcare or become dependent on decisions made elsewhere.