For many in Africa, the questions around AI have largely centered on access – whether the continent has the infrastructure, the skills, or the funding to catch up with the rest of the world. AI has been framed as a distant promise, one that could revolutionize industries, improve public services, and unlock new economic opportunities.
The prevailing concern has often been about the compute power needed to fuel AI’s growth on the continent, whether there’s enough infrastructure in place, and how much more is required to make it happen.
But founder and director of Alliance4AI, Alex Tsado, believes this framing misses the bigger picture. He suggests that the real question isn’t about compute demand at all. Instead, the focus should shift to how AI will transform Africa for the better.
In a keynote address at the Global AI Summit for Africa on the theme “Building Africa’s Compute Capacity—Sustainably and Equitably”, which was held in Kigali, Rwanda on April 3rd and 4th, Tsado challenged leaders and stakeholders to rethink the conversation.
“The biggest question so far on most people’s lips is, what is the compute demand in Africa?” he said. “From where we look at this… this might not entirely be the right question to be asking.”
Drawing from his experience working with NVIDIA’s visionary CEO Jensen Huang back in 2016, Tsado shared that their team didn’t begin by calculating AI demand across the world.
“We weren’t asking, what was the demand for AI in the world? Instead, he’ll ask, how do we show the world that AI is going to change it for the better?”
This shift in mindset, from chasing compute numbers to demonstrating real-world impact, shapes how Alliance4AI is thinking about Africa’s role in the future of AI. Rather than passively waiting for AI to arrive, the focus should be on creating ecosystems that let AI grow with African values at the center.
Tsado walked the audience through a global playbook for driving AI adoption, built on NVIDIA’s experience: seed funding, GPU access, curriculum development, innovation hubs, and model-building support. It’s an approach that worked in the U.S., and it’s one he believes can work in Africa, too.
“One of the most effective ways to support early AI innovation is by providing the foundational tools: computing power, business mentorship, and strategic partnerships,” he explained.
He highlighted the example of InstaDeep, a Tunisian AI startup that began small but scaled through smart partnerships and early access to NVIDIA’s ecosystem. Eventually, the company was acquired for over $600 million.
And now, Africa is seeing its own movement take shape.
Tsado applauded recent steps taken by African players like Cassava Technologies for investing in GPU clusters and building infrastructure that supports homegrown development.
“That’s what kicks off ecosystem growth,” he said. “We can start to see more Instadeeps—maybe 15 of them—emerging from different countries.”
Alliance4AI envisions a pipeline that takes early-stage startups and AI researchers through a support engine, from compute access to data aggregation, incubation, and mentorship. However, there’s also a uniquely African layer: civil society.
“We believe in being better parents to AI than the rest of the world has been,” he said. “We can build AI with African values and show what inclusive, people-first innovation looks like.”
With an initial $50 million commitment to energize this engine, Tsado says the goal is to power 200 startups and prove that African AI isn’t just possible, it’s inevitable.
“We’re not just trying to catch up,” he added. “We’re trying to lead differently.”
