WHO’s EIOS 2.0 Brings AI to Early Outbreak Detection
The World Health Organization (WHO) launched an upgrade to its Epidemic Intelligence from Open Sources (EIOS) in October 2025. Smarter and more inclusive, WHO’s EIOS 2.0 is expected to considerably amplify the early warning system’s capabilities. The goal is to prevent or reduce the number and degree of public health emergencies.
Like its predecessor, EIOS 2.0 is a sophisticated web-based interface that sifts through readily available information from various sources, including media, social platforms, official government websites, news sites and other sources. It analyzes the data obtained to identify clues that point toward the possible spread of contagious diseases or public health threats. According to a press release, as of October 2025, 110 nations have joined the initiative.
Ways EIOS 2.0 Benefits Low- And Middle-Income Countries
- Free Access: Member nations and eligible organizations can utilize the WHO’s outbreak detection tool. This is the same technology high-income countries enjoy, at no charge.
- Preventing Economic Fallout: Pandemics impose a heavy burden on families and societies. During the COVID-19 emergency, an estimated 70 million people around the world fell into extreme poverty.
- Minimizing Loss of Life: A model studying the health emergency that led to the COVID-19 pandemic estimated that up to 35% of the deaths in Wuhan, China, might have been avoided had steps to control the disease’s spread been taken one week earlier. In other cities, 50% of those who perished might have been saved.
- Interface Translations: EIOS 2.0 is now available in multiple languages, increasing accessibility for users with limited English proficiency.
- Semantic Search: Keyword search is now smarter as the system identifies context and intended meaning.
- Radio Sources: Enabling a speech-to-text feature allows EIOS 2.0 to investigate radio communications. This could potentially catch information that may not have been otherwise captured, particularly in more vulnerable areas.
Does Epidemic Intelligence Work?
Africa experiences the highest number of health emergencies each year. Indeed, as of November 2023, there were 130 active outbreaks across the continent. In the same year, an evaluation of the EIOS system showed that in 22 countries, 50% of health events were detected before national announcements or official WHO communications.
WHO studies show that in countries using EIOS, the median time between the first detected health event and notification to the Pan American Health Organization (PAHO) or WHO dropped from 14 days to 11. In fact, it was the EIOS system that first detected a “pneumonia” outbreak in Wuhan, China, in 2019.
One example is the Korea Disease Control and Prevention Agency (KDCA), which utilized the EIOS system between June and October 2023 to detect 425 events. These mainly included mpox and dengue fever. Eight of these events were identified before official declarations, on average, 20 days earlier. In Brazil, an event involving Haff disease was identified and tracked in 2022.
EIOS 2.0’s Promise for Inclusive Global Health Preparedness
In its first iteration of EIOS, WHO demonstrated the advantages of using an intelligent agent that can filter through hundreds of thousands of data sources and discern the likelihood of a health emergency in real-time. Indeed, with EIOS 2.0’s new features, AI capabilities and interface, more regions around the world can benefit from a free-to-use system that can strengthen existing pipelines. With earlier warnings, health authorities can take immediate measures to prevent catastrophic pandemics like those that have sunk millions of people deeper into poverty.
– Johanna Lorena Arredondo González
Johanna is based in Pittsburgh, PA, USA and focuses on Technology and Global Health for The Borgen Project.