AI tool exposes aflatoxin threat in Africa’s maize

AFRICA – An artificial intelligence tool is helping Africa confront a silent danger in its food supply by predicting where aflatoxin contamination is most likely to occur in maize fields.

The Aflatoxin risk Early Warning System (A-EWS), developed by the International Institute of Tropical Agriculture (IITA) and partners through the CGIAR Scaling for Impact and Sustainable Farming programs, will be presented at the 11th African Grain Trade Summit in Zanzibar.

Aflatoxin, a toxic substance produced by the Aspergillus flavus fungus, contaminates maize, sorghum, and groundnuts.

It is linked to stunted growth in children, immune system problems, and liver cancer. Economically, Africa loses an estimated 670 million dollars every year because contaminated grain cannot enter markets.

“There is limited information on the spatial distribution of aflatoxin risk at the farm level, though it is of critical importance for guiding spatial targeting of appropriate management interventions,” explain the researchers behind the system.

The A-EWS uses satellite data and artificial intelligence to create maps that predict contamination risk zones. The system is trained on more than a decade of maize samples collected across Kenya, Uganda, Malawi, and Tanzania, combined with data on rainfall, temperature, elevation, and soil properties.

The model classifies risk into low, medium, and high categories. Tests showed that it performs most accurately when predicting low-risk zones, with slightly lower accuracy for medium and high-risk areas. Rainfall and temperature in March, as well as altitude, are key factors that shape contamination patterns.

“Using these maps… we can finally make aflatoxin visible,” said Jane Kamau of IITA. She added that the maps give policymakers “clear evidence to take action,” replacing broad, unfocused recommendations with targeted interventions.

Practical benefits for farmers and policymakers

The East Africa Grain Council, a partner in the project, described the tool as a “useful platform for the grains industry.” It can help buyers choose safer maize sources and guide farmers on solutions such as the Aflasafe biocontrol product.

Dr. Francis Muthoni, a lead scientist on the project, said, “Aflatoxin contamination has long been an invisible threat, difficult to monitor and manage at scale. Our AI-driven system changes that by providing clear, actionable risk maps that empower farmers, traders, and policymakers to act proactively rather than reactively.”

He noted that the model reflects the complexity of contamination risk. “By integrating diverse data sources, satellite imagery, weather patterns, and historical contamination records, we’ve created a dynamic model that reflects the complexity of aflatoxin risk. This precision enables more effective targeting of interventions, saving crops, protecting health, and preserving market integrity.”

Expanding use and future plans

The team stresses that the accuracy of the system depends on high-quality testing data. Work is underway to create a standard protocol led by Dr. Nancy Kirimi of the Kenya Agricultural and Livestock Research Organization. This protocol will ensure consistent data collection across regions.

Future plans include allowing partners to upload new data to refine predictions. The system may also expand to cover other vulnerable crops like groundnuts and sorghum.

With African governments, research bodies, and industry leaders preparing to meet in Zanzibar, the message is clear: partnerships are essential to roll out innovations that protect food systems and public health.

The interactive dashboard with aflatoxin risk maps is available online at: https://iita-uuwp2.shinyapps.io/Predicted_MaizeAfla_SEA/

Sign up HERE to receive our email newsletters with the latest news updates and insights from Africa and the World, and follow us on our WhatsApp channel for updates.