Kenya's adoption of artificial intelligence (AI), a new-age technology, is focused on the agriculture sector, where machine learning is being used to provide data-driven advice to local farmers to help optimise productivity, a new report has found.
of AI for Africa According to the report published by the Global Systems for Mobile Communications Association (GSMA), an association of mobile network operators, agriculture and food security accounted for 49% of total AI adoption, followed by climate change mitigation and energy use at 26% and 24%, respectively.
The majority of use cases fell into the predictive AI category, according to the report published on Tuesday, which the GSMA said was due to several factors, including the availability of historical datasets, ease of application and lower computational requirements compared to generative AI models.
“In Kenya, where agriculture continues to play a key role in the economy, AI innovation is most prevalent in the agritech sector. AI is already being used in agricultural advisory and financial services, and companies such as Apollo Agriculture are developing alternative credit scoring methods,” the report said.
For example, Microsoft's AI for Good Lab has developed a spatio-temporal machine learning model to detect malnutrition hotspots, enable timely interventions and targeted assistance, and ultimately reduce the impact of malnutrition on vulnerable populations.
The report noted that increasing investment in data centers by the country's leading technology companies and mobile network operators (MNOs) is a key factor driving the push to bring significant storage and computing capabilities to the local level.
In the climate change arena, the use of AI in biodiversity monitoring and wildlife conservation is becoming more prominent, led by large tech companies like AI for Good Lab and nonprofits like Rainforest Connection.
However, the GSMA notes that there is a risk that AI will exacerbate existing socio-economic inequalities, identifying gaps in critical infrastructure and regular power outages as obstacles that perpetuate the digital divide and disproportionately affect lower-income, less-educated and rural residents.
Another major barrier to AI adoption cited in the report is the high cost of hardware such as graphics processing units (GPUs) and cloud computing, particularly for local entrepreneurs and researchers with limited financial resources.
GPUs and cloud computing systems are enablers that provide the storage capacity and computing power required to process complex algorithms, analyze large data sets, and run advanced AI applications and models.
According to the report, the price of GPUs in Kenya is equivalent to 75% of GDP per capita, making it 31 times more expensive than in high-income countries.
“A large skills gap is also hindering the development of the AI ecosystem and use cases. Although universities offer AI-related courses, they often cannot keep up with industry needs, and students have limited opportunities for practical learning and hands-on experience,” the association wrote.
“There has also been an excessive emphasis on core AI skills, such as machine learning and data science, and too little emphasis on building the interdisciplinary skillsets needed to leverage AI to address pressing socio-economic challenges.”
Domestically, deep-tech startup Fastagger is developing software infrastructure that allows machine learning and AI models to run directly on edge devices such as low-cost smartphones.
