WHO Calls AI a Turning Point for Global Health Policy - and a Serious Threat

The World Health Organization has entered the AI debate with unusual directness. A new discussion paper published in June 2026 - titled Artificial Intelligence and Evidence-Informed Policy - makes the case that AI is fundamentally reshaping how health policy is made, and that the stakes of getting it wrong are enormous.

"The policy conversation on AI has focused on clinical care," the WHO paper notes. But the deeper transformation, the organisation argues, is happening at the level of governance itself - in how evidence is gathered, synthesised, and used to make decisions that affect the health of billions.

The Opportunity

AI can process volumes of health data that no human analyst could handle. Surveillance systems can detect outbreak patterns weeks earlier. Literature synthesis that once took research teams months can be done in hours. Policy modelling that required supercomputers a decade ago can now run on cloud infrastructure.

For lower-income countries with limited research capacity, AI offers the prospect of evidence-informed policymaking that was previously out of reach.

The Threat

But the WHO paper is not a technology endorsement. It raises fundamental concerns about what happens when AI-generated evidence becomes the input for high-stakes health decisions - especially when the models producing that evidence are opaque, trained on data that does not reflect the populations being governed, or controlled by private companies with commercial interests.

The risk is not that AI produces wrong answers. The risk is that it produces convincing wrong answers at scale - and that policymakers, under time pressure and with limited technical literacy, do not have the tools to recognise when that is happening.

Geopolitics compounds the challenge. WHO financing reform, Pandemic Agreement implementation, and UN restructuring are all unfolding amid sovereignty tensions and funding uncertainty. In this environment, multilateral platforms face growing difficulty sustaining the global public goods - outbreak surveillance, R&D coordination, emergency response - that depend on shared, trusted evidence.

The Bottom Line

AI in health policy is not coming. It is here. The WHO paper is a call for the field to catch up - to build governance frameworks, audit mechanisms, and international standards before the technology becomes too embedded to challenge.

The health of populations in the next decade may depend on whether that call is answered.