African organisations need to clean up their data and remove friction between technology systems if they are to ensure growing artificial intelligence (AI) adoption results in meaningful business value.
This is according to Karthik Akula, regional head, Middle East and Africa at Freshworks, speaking to ITWeb Africa at the ITWeb Cloud Summit 2026 in Johannesburg.
Akula said the next phase of AI adoption will be less about waiting for more capable AI models and more about preparing organisations to use the technology effectively across existing systems.
“The difference is going to be how to cut through the clutter in organisations, which impedes AI adoption.
Cutting through this clutter means asking and answering fundamental questions Akula added, such as can you clean up your data?
"Can you try to remove the friction points between systems to enable AI to operate amongst multiple connected systems and get the context that is necessary for it to take some actions?”
To be action-ready, AI must be able to operate within the permissions, policies and governance frameworks established by organisations.
However, Akula said AI adoption remains uneven across the continent, reflecting differences in IT maturity, industries and regulatory environments.
For organisations that can overcome those internal barriers, AI could provide an opportunity to help narrow the technology gap between African businesses and more mature international markets.
“AI is a fantastic opportunity for African organisations to really level up,” Akula said.
Rather than requiring African businesses to replicate the infrastructure built by more mature technology markets, AI could enable organisations to build on capabilities already available through cloud-based software and platforms.
Africa’s infrastructure constraints, including connectivity and energy challenges, do not necessarily prevent organisations from benefiting from AI, Akula said.
The growing availability of AI through software-as-a-service and platform-as-a-service models means organisations do not have to establish large physical AI infrastructures locally.
At the same time, the growing availability of large language models means organisations can focus on developing applications tailored to local requirements rather than building the underlying AI technology themselves.
“Africa can benefit from building and customising specific use cases that are relevant to the continent.”
However, as Akula explains, the challenge is to ensure those applications are supported by the organisational foundations needed to make AI useful.
For Akula, that means focusing on adopting technology that is already available rather than waiting for a future generation of AI to solve organisational shortcomings.
“The real value comes from its adoption, even in the current form, rather than looking for a better version of AI,” he concluded.
Share