AI helps African farmers adapt to climate change

In Matungulu ward in Kenya’s Machakos County, John Wambua opens a smartphone app, photographs his window-paned maize leaves and receives a diagnosis within 20 seconds. The problem is a fall armyworm infestation, and the app recommends possible solutions to address it.

“This app has greatly helped me and many farmers around this area,” he said, adding the app diagnoses problems, provides instant responses, tailored recommendations and expected outcomes.

Wambua said he no longer has to call an agronomist to test his soil pH, as the app performs the task, helping him cut costs. Soil pH is a measure of soil acidity or alkalinity.

“The app also provides weather forecast, enabling me to make informed decisions such as when to plant, weed or spray my crops,” he said. “When the rains started, many farmers did not plant because they assumed it was not yet the onset of rainy season. I planted because the app had predicted the rains had begun. That is why the size of my maize crop is bigger than the rest in the village.”

Cynthia Ayekoh, another farmer from Kangundo, a town in Machakos County, expressed gratitude for the innovation, describing it as both innovative and revolutionary.

“I no longer have to go looking for an agronomist. I just use the app and get a solution right away, which is also cheaper for me,” she said.

Ayekoh said the app has helped her detect pests and diseases early and before they spread to worrying levels. She said the tool helps avoid misdiagnosis, which previously led to farmers spraying wrong chemicals and damaging crops and yields.

She hopes the company will introduce an offline version and local language options to increase accessibility, noting that many farmers have limited formal education.

Both Wambua and Ayekoh use PlantVillage+ app, which was introduced in the area about three years ago.

In Kenya, and across Africa, AI is quietly being adopted in many sectors, primarily agriculture, the dominant sector in many African countries. In addition to PlantVillage, other tools such as Virtual Agronomist have also been used by Kenya’s smallholder farmers to get real-time advice on pest control, fertilizer usage, and crop disease detection.

Musau Mutisya, a sales representative for PlantVillage+, based in Machakos County, said the app is accessible through Google Play Store. He said the company is currently developing software for commercial farms.

“The software will help farmers manage activities on their farms, such as spotting infestations, determining when to apply fertilizer and identifying when crops are ready for harvesting,” he said.

Mutisya said he is currently working with more than 400 farmers in the county.

Mercyline Tata, brand manager at PlantVillage+, said the application is available in more than 40 countries worldwide, with a strong presence in Africa.

“Every month we have about 400,000 to 500,000 users on our platform,” she said.

Raphael Ntonja, a machine learning engineer at PlantVillage+, said the company trains its AI model using several crop images showing different diseases.

“We photograph various crops and diseases, collect the data and label the diseases on the leaves before training the model,” he said. “After training, we evaluate the model ourselves before deploying it for farmers to use.”

Ntonja said AI has the potential to transform agriculture and strengthen food security by decentralizing expertise.

“The knowledge that farmers previously needed to get from an expert can now be accessed through AI,” he said.

“No matter where a farmer is, including in the most remote areas, they can get help through a phone. It will also support climate-smart agriculture by sharing daily weather reports so farmers can better plan to cultivate, as well as track carbon.”

Harun Katusya, a data scientist and chief executive officer of the Africa’s Premier AI Conference 2026, said artificial intelligence is poised to transform Africa’s agriculture from reactive, survival-based farming into predictive, precision-driven agribusiness.

He said that over the next decade, AI will guide farmers on when to plant, irrigate, fertilize, and harvest using data from satellites, sensors and weather systems, dramatically improving yields and efficiency.

Real-time advice

“Through mobile-based AI tools, including WhatsApp bots, even smallholder farmers will be able to access real-time agronomic advice, replacing the shortage of extension officers,” Katusya said.

He said that AI will optimize logistics, storage and market access, reducing postharvest losses and improving price discovery.

“AI will turn agriculture into a data economy, where decisions are informed, risks are reduced, and productivity becomes predictable,” he said.

Katusya noted that AI will play a critical role in improving food security across the continent by boosting yields through optimized inputs and crop selection.

It will also reduce losses through early pest and disease detection, improve resource use such as water, fertilizer, and land, and strengthen distribution systems through supply chain intelligence.

He said that AI advisory tools are already helping farmers adapt crops and farming practices in real time during shocks such as droughts and floods.

“AI is not just about producing more food; it is about producing smarter, preserving more and improving distribution,” Katusya said.

Speaking at a webinar convened by Kenya’s Strathmore Agri-Food Innovation Center in March, Joseph Gitonga, the center’s principal lead AI for agriculture transformation, said AI delivers the greatest value when it augments existing human systems rather than replacing them.

“Diagnostic and advisory AI tools are currently the most practical applications because they provide measurable and immediate outcomes,” he said.

Elizabeth Wamicha, an AI and digital innovation researcher and advisor at Qhala, a Nairobi-based digital innovation firm, said trust in AI systems depends on transparency in how farmer data is collected, stored, and used.

“Farmer data literacy is critical so producers understand the value and risks associated with sharing their information. Farmers should be treated as knowledge contributors rather than passive data sources. AI development must shift from data extraction toward farmer empowerment and cocreation,” she said.