Can You Use AI Against Infectious Diseases Without Coding?
How the AI4PEP team in Brazil is making AI tools usable without programming expertise
Infectious diseases can cause epidemics and pandemics, and outbreaks are still hard to predict. Artificial intelligence (AI) tools can help track how diseases spread, but building them usually requires programming skills that are not common in many biology and research fields.
When Skills Become a Barrier
The skills needed to build AI tools, such as machine learning (ML) and data science tools, can create inequalities. Two examples are:
- Accessibility inequality: people and institutions without these skills cannot benefit from AI tools in the same way. Smaller institutions and regions with fewer resources are often left behind.
- Knowledge inequality: ML is complex, and the skills it requires hold back new research.
A 2025 survey asked people involved in health data science, including doctors, academics and epidemiologists, from 54 low- and middle-income countries about the barriers they face. Brazil had the most respondents. Lack of data science skills and training was the most common barrier (33% of comments), followed by funding (20%). As a result, there is a risk that the benefits of AI stay mainly with the institutions that have the skills and resources, rather than reaching everyone working to fight infectious diseases.
A Platform for Non-Experts
To reduce these inequalities, the Brazil team of the Artificial Intelligence for Pandemic and Epidemic Preparedness and Response (AI4PEP) network is building an integrated, user-friendly platform called AutoAI-Pandemics. The team is based at the Universidade de São Paulo in São Carlos and funded by Canada’s International Development Research Centre (IDRC). The platform is designed to let non-experts, such as biologists, physicians and epidemiologists, use data science and ML without needing programming expertise.
The platform focuses on three areas:
- Tracking and predicting outbreaks. It is designed to analyse past disease data automatically, to follow how a disease spreads and what impact it has, and to predict what might happen next. It could also suggest which interventions might work best, saving researchers’ time.
- Studying what causes disease. It analyses biological data, such as gene sequences, to help researchers study pathogens (the viruses, bacteria and other organisms that cause disease) and their variants.
- Fighting misinformation. It includes tools that help people check whether what they read is accurate.
Prof. Dr. Robson Bonidia, a member of the AI4PEP team in Brazil, explains what this means for the team:
“For us, AI for all means that access to this technology must not depend on programming skills, geographical location, social group, race, sex, gender, language, or culture. Instead, we envision AI developed by our own people, grounded in our reality, and designed to meet our needs, ensuring digital sovereignty. AI that gives everyone the opportunity to contribute to this discussion, that is representative, and that understands our differences and local contexts. But above all, it means ensuring that everyone–whether in major urban centers, small towns, Indigenous communities, Quilombola communities, riverside communities, urban peripheries, or rural areas–has access to AI, along with the knowledge and tools needed to develop and use this technology ethically, responsibly, and effectively.”
Tools for Researchers and the Public
The AutoAI-Pandemics tools are built for different users, from researchers studying pathogens to members of the public looking for reliable information.
- For researchers, the team built BioAutoML. Imagine Carla, a biologist with hundreds of gene sequences, the genetic code of a pathogen she is studying. She wants to sort them into groups using machine learning, but that normally requires programming skills. Instead of getting stuck behind a technical wall, Carla uses BioAutoML. She gives it examples of sequences whose groups she already knows, and the tool builds an AI that can sort the rest for her.
- Members of the public can turn to Dominique. Now picture José, who receives a message claiming that a new pandemic has started. Instead of passing it on, he opens Dominique, an AI assistant built by the team, through a chatbot on Telegram and sends it the message. Dominique then helps him judge whether the message is true or not. Alongside Dominique, the team has launched S.N.A.K.E., which goes a step further. It works like a digital tutor: through conversation, it explains how it analyses a piece of news, helping people learn to recognise misinformation.
AI Beyond the Experts
Epidemics and pandemics will keep testing health systems. What the AI4PEP team in Brazil shows is that the skills needed to use AI do not have to stay with a few experts. They can be built into tools that give biologists, physicians and the public the capacity to use AI themselves.
If you work with biological data, or simply want to check the information you receive, explore the team’s tools on the AutoAI-Pandemics website and share them with colleagues.