For 69-year-old Mulu and her extended family, survival depends entirely on her small farm in Kitui County. With seven children and 25 grandchildren to feed, she has long relied on the land to provide for their daily meals. But as rainfall patterns continue to change, her once-reliable source of food is no longer enough.
According to the African Medical and Research Foundation (AMREF), Kitui is one of six counties in Kenya experiencing increasingly shortened rainy seasons. This has led to dwindling crop yields, forcing Mulu and her family to make difficult choices about food consumption. With no other source of income, she has at times relied on food donations, but these are becoming scarce.
Mulu now fears that her family is on the brink of a full-blown drought, reducing them to eating just one meal per day, often nothing more than porridge. The uncertainty of when or if they will have their next proper meal weighs heavily on her. She expresses her distress, saying, “Now we are facing a drought, and we don’t know where to get food. Some time back, we used to get food aid, but nowadays, it’s no longer coming. Now I don’t know whether we will trim our stomachs because we don’t have any more food provisions. We will be taking porridge in the morning and waiting for a big meal in the evening or the following day.”
For Mulu and countless others across Kenya’s arid regions, the struggle for food is an unending cycle of uncertainty. She dreams of a solution that would end hunger for good, saying, “If you can do research and invent a technology to end hunger, I will personally be very happy. I will forever be thankful because you have discovered what will end the hunger that is killing people and their livestock.”
Now, a team of researchers from the University of California, in collaboration with Microsoft, is working on a breakthrough artificial intelligence (AI) tool that could provide such a solution. Their goal is to help vulnerable communities prepare for crop failures and prevent cases of malnutrition before they reach crisis levels.
The urgency of their work is underscored by the devastating impact of the 2023 drought in the Horn of Africa, which hit Kenya hard. According to the United Nations’ World Food Programme, 4.4 million people in Kenya are suffering from acute food insecurity, with approximately 1.1 million women and children facing acute malnutrition.
To combat this growing crisis, researchers have collected extensive data, including detailed weather patterns, NASA satellite imagery showing vegetation changes, and clinical health records spanning the past decade. By feeding this information into sophisticated computational models, the AI tool can predict areas most at risk of food insecurity months in advance.
Since February last year, AMREF has been trialing the AI model in Kenya, and early results show promising accuracy in predicting food shortages three to six months ahead of time. While precise figures on its effectiveness are not yet available, the technology is already shaping a more proactive approach to fighting hunger.
Samuel Mburu, who is helping to develop the AI software for AMREF, highlights its potential impact. “We have looked at historical data for the last 10 years, from different data sources, to define the number of cases all the way to the sub-county level. The aim is to help allocate resources ahead of time so that we can intervene for children under the age of five, ensuring they receive the nutrition they need, especially at our health facilities.”
The ability to predict food shortages before they happen allows humanitarian organizations and local governments to take preventive measures, rather than simply reacting to crises as they unfold. AMREF hopes that by alerting communities early, families like Mulu’s can better prepare for lean periods, securing resources in advance rather than waiting until hunger takes hold.
Initially focused on six northern and eastern counties most prone to drought, the AI model has now expanded its coverage to include the entire country. One of the key challenges researchers faced was that health records alone were not always accurate enough to predict food insecurity, as many people in remote areas lack access to medical facilities. To improve data accuracy, AMREF plans to incorporate community-level information into the model, making it even more reliable.
Mburu emphasizes the power of data in changing the way hunger is addressed. “We now have the ability to ingest vast amounts of data, analyze it, and bring real value to it. In the past, data was primarily used for historical reporting, but now we are using it for prediction, allowing us to take preemptive action. Another key aspect is resource mobilization. If we can build a compelling case with strong data, we can source for resources both internally and externally, ensuring that aid is directed where it’s needed most. The main challenge, however, is cost—especially with cloud-based AI environments. Fortunately, we received generous sponsorship from Microsoft for this particular initiative, but long-term sustainability remains a concern.”
The technology also aims to assist county officials in making better preparations for food insecurity. Mburu explains, “Traditionally, cases of malnutrition would be reported after they occurred, and resources would be provided reactively, such as food packs for children already suffering from malnutrition. With the new technology, we can predict malnutrition levels three to six months in advance and mobilize resources beforehand. This ensures that communities receive help before conditions become critical, preventing suffering rather than responding to it.”
For families like Mulu’s, this kind of foresight could mean the difference between resilience and catastrophe. While no single solution can completely eliminate hunger, the AI model represents a significant step toward smarter, data-driven interventions. By leveraging technology to anticipate crises rather than simply reacting to them, Kenya is moving toward a future where fewer families will be forced to endure the heartbreak of watching their children go hungry.
Still, challenges remain. The high costs associated with AI technology, the need for sustainable funding, and ensuring that predictive models are accessible to the most vulnerable communities all require ongoing attention. However, with continued investment and collaboration, AI-driven solutions have the potential to transform how food insecurity is tackled—not just in Kenya, but in other regions facing similar struggles.
For now, Mulu holds onto hope that something will change. “If there is a way to stop hunger, we need it now,” she says. “We cannot continue to wait.”
With AI offering the promise of early intervention, Kenya may finally be on the path to breaking the cycle of food insecurity, bringing hope to millions who have long faced the uncertainty of when—or if—their next meal will come.
