Africa’s AI ambitions will be decided less by the power of emerging models than by whether organisations on the continent can put the right internal structures, data standards and governance frameworks in place. That is the assessment of Karthik Akula, regional head for the Middle East and Africa at Freshworks, who spoke to ITWeb Africa at the ITWeb Cloud Summit 2026 in Johannesburg.
Akula’s central argument is directed at decision-makers within organisations rather than at the technology itself. The next phase of AI adoption, he said, will be less about waiting for more capable models and more about preparing organisations to use the technology effectively across their existing systems. In his framing, the burden of responsibility sits with leadership to confront questions that are fundamentally organisational: can the data be cleaned up, and can friction points between systems be removed so that AI can operate across multiple connected systems and gather the context it needs to take action?
“The difference is going to be how to cut through the clutter in organisations, which impedes AI adoption,” Akula said.
That clutter, in his account, is not merely technical. For AI to be action-ready, it must operate within the permissions, policies and governance frameworks that organisations themselves establish. Where those frameworks are absent or inconsistent, the technology cannot function as intended, regardless of its underlying capability.
Adoption across the continent, Akula noted, remains uneven, reflecting differences in IT maturity, industries and regulatory environments. Those variations mean the accountability for progress rests with individual organisations and the regulatory conditions in which they operate, rather than with a single continental trajectory.
For organisations able to overcome those internal barriers, Akula sees an opportunity to narrow the technology gap between African businesses and more mature international markets. “AI is a fantastic opportunity for African organisations to really level up,” he said.
Importantly, he argued, this does not require African businesses to replicate the infrastructure built by more mature technology markets. AI could allow organisations to build on capabilities already available through cloud-based software and platforms. 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, a point he made even as he acknowledged Africa’s infrastructure constraints, including connectivity and energy challenges. Those constraints, he said, do not necessarily prevent organisations from benefiting from AI.
Meanwhile, the growing availability of large language models shifts the emphasis of decision-making. 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,” Akula said.
Yet the challenge, as he explained it, is to ensure those applications are supported by the organisational foundations needed to make AI useful. In other words, the availability of powerful tools does not substitute for the internal work of governance, data quality and system integration. That work falls to the institutions and leadership teams responsible for how technology is deployed within their mandates.
Akula’s concluding message was directed at organisations tempted to defer investment until a more advanced generation of AI arrives. For him, the priority is adopting technology that is already available rather than waiting for a future model 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.
Taken together, his remarks place the responsibility for Africa’s AI outcomes squarely on organisational readiness: clean data, reduced friction between systems, and clear governance frameworks are the conditions on which meaningful business value from AI depends. The open question for the continent’s leadership teams is whether they will build those foundations now or wait for a future model to solve shortcomings that technology alone cannot fix.