The 53 Member States of the World Health Organization’s European Region unanimously adopted the region’s first Roadmap for Action on Artificial Intelligence in Health during a meeting in Copenhagen on October 29. The plan covers the period from 2026 to 2033 and sets out priorities for integrating AI into medical services while ensuring that its use remains safe, effective and accessible.
The agreement comes as AI tools are becoming more common in health care, from systems that help detect diseases to applications that support clinical decisions and handle administrative tasks.
A WHO/Europe assessment published in 2025 found that nearly two-thirds of countries in the region were already using AI for diagnosis, while half had introduced conversational AI tools. However, only one in five countries included AI in medical education, and just one in four provided training for health workers already in service. Fewer than half had assessed whether their existing laws covered AI in health care.
Read also : Seven candidates nominated for WHO director-general position
Four priorities for AI in health systems
The roadmap identifies four areas for action: governance and leadership; ethical, regulatory and legal frameworks; data and digital infrastructure; and capacity-building, implementation support and learning.
It calls on governments to incorporate AI into national health priorities, assign clear responsibilities to public institutions and provide sustainable funding. It also stresses the need for safeguards throughout the development and use of AI systems.
Data quality, secure access, interoperability and cybersecurity are among the foundations identified as necessary for reliable AI applications. The plan also calls for investment in workforce skills and the evaluation of AI tools in real-world health-care settings.
Dr Corinne Capuano, Officer-in-Charge for WHO/Europe, said AI could help identify diseases earlier, alert medical staff when patients’ conditions deteriorate and reduce administrative work for clinicians. It could also give people access to initial information about symptoms at any time.
She cautioned, however, that governments must address questions of safety, fairness, patients’ rights and public trust.
Dr Natasha Azzopardi-Muscat, WHO/Europe’s Director of Health Systems, said AI could support more tailored and timely care when it is properly developed and regulated. She stressed that human interaction would remain central to health services.
Read also : WHO sets 2030 target for countries to detect health threats within 24 hours
New WHO support for implementation
The roadmap is accompanied by an implementation guide designed to help countries with different levels of digital development put the recommendations into practice.
Member States also backed the establishment of a new WHO office on artificial intelligence. It is intended to provide technical assistance, support research and evidence-gathering, strengthen national capacity and facilitate the exchange of experience between countries.
The framework is aligned with existing European and international rules and initiatives, including the European Union’s AI Act, the European Health Data Space and the General Data Protection Regulation. It also draws on guidance from the Organisation for Economic Co-operation and Development.
WHO/Europe said the roadmap builds on its first region-wide assessment of countries’ readiness to use AI in health, published in 2025, which gathered responses from 50 of the 53 Member States.
A midterm report is scheduled for 2029 to assess progress in implementing the plan.
