Industry leaders have been cautioned against expectations of quick fixes from Artificial Intelligence (AI) despite a heated race to adopt the technology as a booster for growth.
AI technology allows computers and machines to simulate human intelligence and problem-solving tasks. The technology can be applied to many sectors, including healthcare, manufacturing, mining, and the military.
Business leaders in Kenya have identified credit risk assessment, cybersecurity, customer service, fraud detection, and marketing as top priority areas for enhanced AI investments.
Questions, however, linger over whether the technology will deliver meaningful returns amid rising costs, data gaps, and growing security concerns.
“AI is not something you deploy today and get returns tomorrow,” Hani Nofal, Senior Vice President for Technology Solutions in the Middle East and Africa at business and technology services firm NTT Data, said.
“Businesses should expect AI return on investment to be long-term. The biggest value comes when you bring AI closer to your core operations, not when you treat it as a side project,” he told Business Daily in an interview.
According to KPMG’s 2025 Africa CEO Outlook Survey, local executives- particularly in financial services, telecoms, and retail- are preparing to increase AI spending over the next year.
Nearly 70 percent of senior executives across the region plan to allocate up to a fifth of their budgets to the technology, a trend mirrored in Kenya’s fast-digitising economy.
But leaders say the real test will be whether AI delivers long-term value, not just quick wins.
Mr Nofal said many firms are still trying to build modern AI systems on infrastructure that was never designed for it.
“A lot of the infrastructure we’ve built for years is not ready for AI,” he said, warning that outdated systems slow adoption and drive up costs.
Security risks are also rising, with companies worried about deepfakes, automated scams, and the possibility that cybercriminals could exploit AI tools faster than regulators can respond.
“For every good use of AI, there is a potentially harmful one,” Nofal said. “We can all fall into a trap if we don’t stay ahead of it.”
Beyond the corporate concerns, public anxiety around job displacement is growing in Kenya. Many see AI as a threat rather than an opportunity. Mr Nofal, however, dismissed fears of full automation, saying AI will streamline work but not erase it.
“AI will augment certain tasks, but the human in the loop will always be required,” he said.
Kenya has singled out healthcare, education, and agriculture as top priority sectors in its new AI deployment strategy.
Other sectors mapped out for early focus include public service delivery, security, the creative sector, sustainability, and small and medium-sized enterprises.
Within the healthcare sector, the Kenyan government is considering the development of a maternal health chatbot in local dialects to provide accurate pregnancy and childbirth information, in addition to an expanded disease advisory system that will build on existing platforms.
Kenya’s AI strategy further notes that in the education sector, priority would go to intelligent tutoring systems and multilingual teacher training modules to improve access and quality of instruction, while emerging use cases in agriculture include translating existing data into farmer-friendly audio formats in local languages and developing AI-powered fertiliser recommendation systems.
“Public-sector use cases include multilingual chatbots and virtual assistants to improve service delivery, while the creative sector could benefit from an AI-powered national digital creative platform to enhance market access and support local content creators,” notes the strategy document.
To this end, the government has identified three pillars to accelerate Kenya’s AI uptake, including modernisation of the national digital infrastructure for AI access and development, the establishment of a robust and sustainable data ecosystem for AI and innovation, as well as incentivising the development of cutting-edge localised AI models and solutions through homegrown research and development.
