The UN Food and Agricultural Organization (FAO) states the following:
Agriculture is key to Kenya's economy, contributing 26 per cent of the Gross Domestic Product (GDP) and another 27 per cent of GDP indirectly through linkages with other sectors. The sector employs more than 40 per cent of the total population and more than 70 per cent of Kenya's rural people.
According to FAO, the sector also accounts for 65 percent of the export earnings, and provides the livelihood (employment, income and food security needs) for more than 80 percent of the Kenyan population and contributes to improving nutrition through production of safe, diverse and nutrient dense foods.
Agriculture is therefore Kenya's main socio-economic activity, and yet perhaps the least computerised.
The 2018 Economic Survey Highlights shows that agricultural production dropped significantly for all the key crops. Sugar, Wheat, Rice, Coffee, Tea & Maize recorded drops in yields over the previous year by 33, 23, 20, 11, seven and six percent, respectively.
There must be all sorts of reasons for the reducing productivity in the agricultural sector that may range from corruption, adverse weather conditions to poor farming practices.
But one factor that rarely comes up is the lack of agricultural extension officers.
Gone are the days, when we used to have agricultural extension services, where agricultural experts would visit farmers and advise them on mundane things like soil fertility, appropriate seedlings and crop diseases.
These extension services would add value to the yields produced per hectare.
Increasing yields per unit farmland is the only way of increasing productivity, without indulging in the more glamorous but corruption-prone projects like the Galana-Kulalu irrigation scheme.
If we have no budgets for extensions services but still need to increase productivity, what options do we have?
The answer lies in Artificial Intelligence in Agriculture.
With increased capabilities of current mobile phones and number of Internet connections, any farmer in the rural Kenya can today access the same benefits of AI in Agriculture as his or her peer in the US or Europe does. For example, when it comes to disease identification, a farmer might have to simply to scan the leaves of the affected crop with their phone camera a software would identify the disease and suggest a treatment regime. The same camera can be pointed to the soil samples and a report on the soil fertility, composition, amount of fertilizer or preferred seedling types can be availed to the farmer instantly.
Additionally, we should not be waiting for crop failure in this day and age of sophisticated but affordable weather prediction systems.
With current AI models, the short, medium and long team weather forecast for every inch of planet earth is already mapped out with a high degree of confidence.
The government should be in a position to use such data to easily advice farmers on when to plant and not to plant, so as to save them from loses occasioned by weather-related crop failure.
Furthermore, there are insurance enterprises that would be interested in cushioning farmers in the event that the predicted harsh weather conditions do come to pass.
It is time the agricultural sector to borrowed from the success of AI in the financial sector as it is the only way to increase our agricultural productivity.