Esther Mulu, 69, and her large family often face food shortages due to failed crops. She lives in Kitui County, Kenya, with her seven children and 25 grandchildren, all depending on her small farm for food.
According to the African Medical and Research Foundation (AMREF), Kitui is one of six counties with shorter rainy seasons. In the past, Mulu has relied on food aid, but now, with drought worsening, her family eats only one meal a day, often just porridge.
To help communities like hers, researchers from the University of California, with support from Microsoft, are developing an AI tool to predict food shortages and prevent malnutrition.
The 2023 drought in the Horn of Africa severely impacted Kenya. The UN’s World Food Programme reports that 4.4 million people face food insecurity, and 1.1 million women and children suffer from malnutrition.
The AI tool analyzes weather patterns, satellite images, and health records from the past ten years to predict food shortages. AMREF has been testing the tool in Kenya since early 2023, and results show it can predict food crises 3–6 months in advance.
Samuel Mburu from AMREF explains that this technology helps allocate resources ahead of time, ensuring young children receive proper nutrition. Initially focused on six counties, the model now covers all of Kenya. Since many people lack access to healthcare, researchers plan to include more local community data to improve accuracy.
AMREF believes this AI tool will help officials prepare for food shortages by predicting malnutrition cases early. With this data, resources like food packs can be delivered in time to support affected communities. However, cost remains a challenge, especially for cloud-based AI systems. Microsoft has funded this project, but more resources are needed for long-term success.
Source: msn
