The future of AI in Africa

The future of AI in Africa

AdobeStock_ AI Automation.png

Following reports that Google intends to fast-track the adoption of artificial intelligence (AI) in Africa, Capacity explores if AI will thrive on the continent.

Reports indicate that Africa is lagging behind on AI. However, the continent is beginning to bridge the gap with several initiatives looking to capitalise on its vast potential.

Enter Google. The American juggernaut has launched its Google for Startups Accelerator which is aimed at supporting African startups looking to use AI to solve local challenges. It is just one of many AI projects sweeping the continent as companies look to gain a foothold of a potential goldmine.

With a growing population of over 1.4 billion people and the majority of those under 30, the continent is ripe for investment.

Fatima Tambajang Head of Developer, Startups & VC Ecosystem for Africa at NVIDIA believes that despite the challenges regarding restricted access to hardware, data and talent, Africa has made significant strides in AI.

“This has been demonstrated by several African countries’ AI strategies,” Tambajang says.

There are currently over 30 AI developer communities in Africa – from countries such as Sudan and Mali to more established markets such as South Africa and Tunisia. Many African universities also offer degrees or specialisations in AI, Tambajang is keen to stress.

Further to this, Africa has over 2,400 AI organisations operating across several industries including health, wellness, fitness, farming, law training and insurance. This includes a rapidly growing ecosystem of startups.

Tambajang also notes that while Africa’s tech boom is often linked to fintech, the largest tech acquisition in Africa is an AI-focused enterprise solutions company, InstaDeep, which was acquired for $682 million in January 2023.

Lagging behind?

Baron Fung, senior director at Dell’Oro Group thinks the continent is still lagging behind in AI and in more mainstream fields such as cloud computing.

This, he adds, is evidenced by the lack of presence of the hyperscalers offering public cloud services in Africa.

The exception would be in South Africa, in which Amazon, Microsoft and Oracle provide public cloud services with Google set to launch new services in the country later this year.

“Today, given the complexity of AI applications and infrastructure, enterprise customers are more likely to use the public cloud to deploy their AI workloads,” Fung says.

“The lack of public cloud services in Africa would mean that enterprises would need to invest in their own AI infrastructure on premises along with a complimentary workforce, which makes AI deployments even more complex and challenging.”

Use Cases

AI is already being used in various sectors across Africa, including enterprise, agritech, healthcare, fintech and edtech.

Tamabajang mentions Intron health, a member of NVIDIA’s Inception programme for startups, which is a Nigerian company that developed Africa’s first speech-to-text AI chat tool capable of understanding 200 African accents.

With contributions from over 7,000 Africans across 13 countries, the tool has helped doctors increase their clinical documentation processing speed by six times.

Proteinea, meanwhile, another Egyptian company leverages deep-learning models to develop biotherapeutic medicines at an unprecedented rate.

“AI will impact almost every industry,” Tambajang says.

“At NVIDIA, we’re particularly excited about the immediate opportunities AI is bringing to the African Healthcare, Smart City and Climate Tech industries.”

Fung references a McKinsey article that indicates the main industries in Africa are agriculture, mining and banking. He believes that there are opportunities in those industries for AI.

“For example, with agriculture and mining, AI models could be used to optimise yields and production,” he says.

In the banking and financial sectors, meanwhile, AI could help automate more customer transactions in the commercial banking space. It could also use AI models to make predictions and recommendations on the market or investments.

A major hurdle for African enterprises to invest in AI is to understand its return on investment, Fung adds.

“It’s sort of a chicken-and-egg scenario,” he says.

“And given the lack of availability of the public cloud to develop AI applications in the early stage, that is a valuable resource enterprise lack that is available in other regions.”

Regulation

Fung believes that AI would need regulation. Data sovereignty has been a key initiative in developed regions and that would require African nations to come together to determine where and how customer data and data generated by AI models should reside.

“If not, there may be potential privacy issues with governments utilising the data against citizens in unintended ways.”

Tambajang echoes that sentiment, agreeing that AI must be regulated. She believes that AI is an important technology that affects the social fabric, norms and safety of people.

“Regulation is helpful for technologies that can do a great deal of social and industrial good, but have the potential to be used in harmful ways.

“It’s important for regulators across Africa and globally to understand existing laws and the potential impact of any proposed legislation and get industry input.”

Tambajang is encouraged by efforts from regulators to not approach AI systems monolithically.

Policymakers must understand the potential impact of regulations on innovation before enacting broad and heavy-handed laws, she says. With that said, Tambajang says history has shown over time that new technology is the engine of economic growth.

“AI will augment many jobs, create even more opportunities, improve productivity and enhance our quality of life across the continent and worldwide.”

Gift this article