Experts have warned that unless pressing infrastructure and affordability challenges are addressed, Artificial Intelligence (AI) will remain the privilege of a few, not the inclusive, transformative tool it’s envisioned to be. This was during a panel discussion titled “Infrastructure: Barriers to Adoption and Building Blocks for the Future,” on the second day of the inaugural Global AI Summit on Africa, which concluded Friday in Kigali. ALSO READ: Too much control stifles AI, experts warn governments Veronica M. Nduva, the Secretary General of the East African Community (EAC), underscored that Africa stands on the brink of AI-driven revolutions, with the potential to redefine innovation, productivity, and service delivery across the continent. She stressed that the continent’s inadequate digital infrastructure continues to hamper its AI ambitions. “For all the promises that AI holds, one undeniable reality remains: Africa’s inadequate digital infrastructure is a major impediment to realizing these ambitions,” she noted. “Despite being home to 60 percent world’s arable lands, vast mineral wealth, and a young, dynamic population, Africa continues to grapple with infrastructural barriers that hinder AI aspirations.” ALSO READ: Africa must not be left behind, Kagame tells AI Summit Among the most significant challenges is a lack of reliable electricity. Nearly 600 million Africans live without stable access to reliable power, Nduva noted, a figure that rises to over 70 percent in rural areas. This remains a challenge in sustaining data centres, computing power, and cloud services that are all essential for AI development. To meet the high energy demands of AI, experts at the summit emphasized that governments and public institutions must collaborate with the private sector to develop sustainable and affordable energy solutions, including solar and nuclear, that can support the scale and complexity of AI technologies. Connectivity remains another pressing challenge. Internet penetration in Africa is approximately 43 percent, which is significantly lower than the global average of 66 percent, according to Nduvu. “The cost of data remains disproportionately high compared to average income. In some African countries, 1GB of mobile data costs as much as 5 percent of an individual’s monthly earnings, compared to the global affordability standard of 2 percent or lower,” Nduva said. She added that without affordable and widespread broadband access, “AI adoption will remain an elite privilege rather than a transformative tool for all.” The cost of devices is also a major limiting factor. According to Nduva, a high-quality smartphone in Africa can cost more than 50 percent of the average monthly income, and a personal computer is even less affordable for most. “These costs create barriers for young innovators, students, and entrepreneurs who seek to engage with AI technologies,” she said. While AI development requires intensive training and research, Nduva stressed that Africa accounts for less than two percent of the world’s AI-related research publications. She attributed this not to lack of talent, but to limited access to the necessary computational resources, mentorship, and funding. To bridge this gap, she called for stronger public-private partnerships to establish AI Centers of Excellence across the continent, spaces where universities, governments, and the private sector can jointly develop local talent, build capacity, and drive innovation. The EAC, she added, has secured $40 million from development partners to build resilient digital infrastructure, strengthen a resilient environment for cross-border data, boost last-mile digital infrastructure, digital capacity building, and innovation. “EAC is working on enacting a regional law to formalize and regulate these data flows, setting the stage for enhanced collaboration and innovation,” she added. Amb. Clever Gatete, the Executive Secretary of the UN Economic Commission for Africa, emphasized the need for regional and continental approaches to infrastructure development. “This is linking country to country, but also the whole continent. I think this is where infrastructure and training become useful,” he said. According to Ghanaian professor and investor Stephen Moore Kwesi, founder of Khaya AI—an African language translation tool—the continent can only benefit from the ongoing technology revolution if its techpreneurs unite to solve Africa’s unique market challenges. Khaya AI tackles the lack of digital platforms in indigenous languages. The tool currently supports instant translation across 11 African languages, allowing farmers, patients, drivers, and consumers to communicate in their native tongues. “This year we are looking to add Kiswahili and Kinyarwanda to our platform as we look to improve accessibility and communication in sectors like agriculture, healthcare, and education,” Kwesi shared. Reacting to the issue of inclusivity, Kwesi is making the Khaya AI API available for purchase by local innovators in Rwanda, as part of the efforts to localize technology and widen its impact. “It empowers local communities by involving them in the development process, ensuring the technology reflects native perspectives and strengthens cultural identity,” he added. On the issue of network infrastructure, Bocar Ba, the CEO of the SAMENA Telecommunications Council, emphasized that AI cannot thrive without robust and resilient telecommunications infrastructure. Ba also said there should be an enabling environment for call infrastructure, which would include setting up data centers and energy. This, he said, would lower the capital investment for the telecommunication companies while investing in call infrastructure. “We won’t have any AI if we don’t have infrastructure, and that infrastructure has to be funded,” Ba said. “We have the investment, financing, and the funding. Three different foundations for unlocking access to capital, which call up on different stakeholders,” he noted. “The private sector usually which invests, needs to have a favourable regulatory environment, the right policies that will foster investment. And the private sector is looking for profitability, sustainability, and predictability.” He noted that to fully leverage AI and build favourable infrastructures, it is important to redefine the role of different stakeholders. He emphasized that governments should champion the approach, stimulate demand through literacy and affordability, and mitigate the risks of investment. “We have to work as a multistakeholder platform with different institutions,” he said.
