Artificial intelligence (AI) can support public health authorities in communicating risks and managing the spread of false information during public health emergencies, according to a new study published by the University of Zurich and WHO/Europe. However, its use must be guided by strong ethical principles to protect public trust and safety.
The study, published in BMJ Global Health, is the first international, multi- and transdisciplinary effort to comprehensively assess the impact of AI on risk communication, community engagement and infodemic management (RCCE-IM). Through a consensus-building process, a panel of 54 experts from 27 countries evaluated the opportunities and challenges of using AI in emergency management.
A previous WHO/Europe study showed that confusion around health information, especially during outbreaks and disasters, when science is evolving, can negatively impact people’s health decisions and hinder protective action. Meanwhile, University of Zurich and WHO/Europe experts found that AI tools have the potential to significantly improve how health authorities tailor messages to specific populations, listen to public concerns in real time, and enhance the reach and inclusivity of health information and advice, particularly in multilingual, multicultural or underserved settings.
However, the study also underscores the risks related to the use of AI, including algorithmic bias, privacy concerns and the potential to worsen health inequalities. If health messages are not well targeted or data is not used carefully, it can unintentionally harm vulnerable communities or contribute to spreading mis- and disinformation.
“We have seen how rapidly false information can spread in emergencies and impact people’s lives. This is one of the major challenges of our times. AI has the potential to help address this effectively by identifying harmful narratives early and targeting relevant and accurate information to diverse audiences. But while the results of this study are encouraging, they are also a reminder to proceed with caution. Innovation should never come at the cost of trust or safety,” said Cristiana Salvi, Regional Adviser for Risk Communication, Community Engagement and Infodemic Management, Health Emergencies at WHO/Europe.
Dr Daniela Mahl, the study’s co-author and researcher at the University of Zurich, added: “AI’s ability to support or undermine public health efforts depends on how it is governed and implemented. The line between innovation and harm is thin, especially in high-stakes emergencies. Our findings emphasize the need for clear principles and strong cross-sector collaboration to ensure AI really strengthens health communications”.
The study outlines guiding principles and actionable steps to facilitate the responsible use of AI in RCCE-IM. These include establishing clear governance frameworks, training health professionals, and promoting inclusivity and accessibility in AI applications. It also calls for an internationally coordinated approach to managing these technologies, in a way that is fair, people-focused, safe and transparent. It recommends enhancing cross-sector collaboration to ensure that AI tools fit diverse cultural, regional and technological contexts.
“This study highlights the potential and the complexity of using AI in risk communication and infodemic management during emergencies. It’s not just a technical upgrade. It requires rethinking how we design, test and scale these interventions. Can we ensure that AI insights translate into timely, actionable and ethical interventions on the ground? This is why we should invest in both innovative and ethical solutions that build public trust, especially in complex and rapidly evolving contexts,” said Dr David Novillo Ortiz, Regional Adviser for Data, Evidence and Digital Health at WHO/Europe.
Regional frameworks underpin the study’s recommendations. WHO/Europe’s new emergency strategy and action plan, known as Preparedness 2.0, puts RCCE-IM at the core of emergency management and calls for stronger digital literacy. Enhancing country capacities to better govern digital transformation in the health sector is a core priority of the Regional Digital Health Action Plan for the WHO European Region 2023–2030 and associated progress report. This includes researching innovation on predictive analytics for better health through AI, with special attention on infodemic management.
