WHO highlights the opportunities, risks and conditions for responsible use of AI

AI in health policy

In June 2026, the World Health Organization published the discussion paper Artificial intelligence and evidence-informed policy: emerging challenges and opportunities. The document concerns the use of artificial intelligence not only in diagnostics, treatment, or clinical work, but also in the development of evidence-informed health policies. This is an important perspective. AI can influence how health systems define problems, how they design interventions, how they assess their effectiveness and how they monitor the effects of implemented policies. In practice, this means that artificial intelligence is beginning to enter the space where system-level decisions, resource allocation, health priorities and reform directions are shaped.

https://www.who.int/news/item/02-06-2026-new-who-discussion-paper-sets-out-opportunities-and-risks-of-ai-in-evidence-informed-health-policy

AI across the entire health policy cycle

WHO structures its analysis around the health policy cycle. It indicates that AI can support several key stages, like problem identification, the design of possible solutions, policy implementation, monitoring of effects and adjustment of actions over time.

At the problem identification stage, AI can help analyse large and diverse datasets. It can support the identification of epidemiological trends, health inequalities, gaps in access to services, or areas where existing interventions do not deliver the expected effects.

At the policy design stage, AI can support scenario modelling, the analysis of potential effects of different intervention options and faster comparison of available choices. In health systems operating under pressure from limited resources, such tools can help make more informed choices about priorities.

At the implementation and monitoring stage, AI can support adaptive policy management. This means the ability to detect more quickly whether a given intervention is working as intended, where deviations are emerging and what corrections are needed. WHO points out that AI can support more responsive and iterative decision-making in complex health contexts.

The most important opportunity- faster use of evidence

One of WHO’s main arguments is that AI can accelerate the analysis, synthesis and use of evidence. In health policy, the problem is often not only a lack of data. The problem is also its fragmentation, varying quality, limited comparability and the slow translation of knowledge into decisions. AI can support the integration of data from multiple sources, literature analysis, the creation of living evidence reviews, predictive modelling and scenario simulation. For decision-makers, this means the possibility of working with a more up-to-date and broader picture of the situation. In digital health practice, this issue is particularly important. Health systems generate more and more data, but the availability of data alone does not guarantee better health policy. What is needed are tools that help distinguish signal from noise, connect quantitative data with context and transform information into decisions.

The biggest risk- the apparent objectivity of AI

WHO shows very clearly that AI in health policy can reinforce existing errors. If input data are incomplete, biased, or do not represent specific social groups, the model may perpetuate a distorted picture of reality.

At the problem definition stage, bias in data may make some health needs remain invisible. At the stage of designing solutions, excessive optimisation of measurable goals may narrow policy to what is easy to count, rather than what is most important for population health. At the implementation stage, digital inequalities, cybersecurity and the ability of institutions to manage technology matter. At the monitoring stage, even subtle errors in AI tools may gradually move policy away from its original goals. This is one of the key conclusions from the WHO document- AI can increase the scale of analysis, but it does not automatically guarantee better decision quality. In healthcare, system-level decisions require interpretation, context, accountability and ethical assessment.

[embed]New WHO discussion paper sets out opportunities and risks of AI in evidence-informed health policy The World Health Organization has published Artificial intelligence and evidence-informed policy - emerging challenges…www.who.int

[embed]Artificial intelligence and evidence-informed policy: emerging challenges and opportunities… Artificial intelligence (AI) is increasingly shaping evidence-informed policy-making (EIP) in health by enabling faster…www.who.int