Spend any time at Enlit Africa, and you’ll hear two letters repeated over and over: AI. The buzz about artificial intelligence – specifically, its applications in the utility space – is impossible to miss; and it’s similarly impossible to not be swept up in the conversation.

Speaking ahead of the 2020 event (back when Enlit Africa was still called African Utility Week, when 2020’s events hadn’t been pushed out to 2021, and COVID-19 didn’t have a name yet), advisory board member Jo Burgess, senior technology specialist at Isle Utilities, spoke of AI’s transformative potential. ‘I think the water sector in Africa is about to meet a massive digital revolution, and I am excited to be here to take part in that,’ she said. ‘New, digital/virtual tech like AI, blockchain and AR/VR will revolutionise the way we manage assets, predict events and monitor situations in real time. Certainly I don’t think I personally have fully grasped what the impacts will be.’

AI is not only capturing the imagination of Africa’s public-works space. Fast forward a year (albeit an historically eventful one) from Burgess’ comments, and organisations across all industries – from the public sector to pharmaceuticals, NPOs and more – are deploying AI and machine learning (ML) to respond to the COVID-19 crisis.

‘We’re seeing three main areas where ML is having an impact right now – scaling customer communications, understanding how COVID-19 spreads, and speeding up research and treatment,’ according to Swami Sivasubramanian, VP of ML at Amazon Web Services (AWS). ‘In South Africa, we’ve seen how providing access to advanced technologies such as AI and ML is vital to stopping the spread of COVID-19 and helping individuals quickly find medical help when they fall ill.’

He cites the example of GovChat, South Africa’s biggest citizen-engagement platform, which launched a COVID-19 chatbot in less than two weeks. That chatbot uses Amazon’s Lex AI service to build conversational interfaces into any application using voice and text – in this case, providing health advice and recommendations on COVID-19 testing.

Meanwhile, the UN Economic Commission for Africa (ECA) announced recently that it would launch an African AI Research Centre in the Republic of the Congo, believed to be the first of its kind on the African continent. ‘This centre will not only develop artificial intelligence in the country, but also at the level of Africa, and also help to tackle some ethical issues around AI,’ economic affairs officer Mactar Seck notes on the ECA website.

The UN estimates that AI could contribute US$15.7 trillion to the global economy by 2030. According to Seck, AI ‘offers more development opportunities to the continent and could significantly contribute to creating more jobs’. He adds that ‘the world is projected to witness a 33% growth in the artificial intelligence economy over the next five years, not to mention that the sector can generate up to US$4.3 billion for the African continent within that time’.

That goes some way to explaining why the AfDB recently approved a US$1 million grant for AI-enabled systems to process customer complaints on behalf of the national banks of Ghana and Rwanda, and the Competition and Consumer Protection Commission of Zambia. The project, which will use multilingual AI chatbots, is expected to improve the tracking of customer complaints made to financial-services providers; build confidence in financial services by strengthening support for marginalised groups; and improve the collection of consistent data for the development of consumer-protection policies.

To help drive this AI evolution, several African countries have launched AI research centres, from South Africa’s Centre for AI Research (which has nodes at the universities of Cape Town, KwaZulu-Natal, North-West, Pretoria and Stellenbosch) to the new AI training centre at Cameroon’s University of Yaoundé, and the AI and Robotics Centre of Excellence at Addis Ababa Science and Technology University in Ethiopia.

So, there is investment interest in AI’s potential in Africa, as well as excitement and expectation. But what does the continent’s AI-enabled future look like? Botswana, for example, sees AI and its related Industry 4.0 technologies propelling the country from an upper-middle-income country to a high-income country by 2036. How, though?

As Oxford Insights points out in its Government AI Readiness Index 2020 report, ‘AI readiness in Africa is […] constrained by lack of an appropriately skilled local workforce’. Quoting a Forbes Insights article, the report notes that AI solutions don’t spring fully formed – they emerge from communities of researchers and entrepreneurs, which require a lot of groundwork to build.

‘The talent that does exist tends to face numerous obstacles, including low visibility within the global community of AI researchers and entrepreneurs, due, for example, to visa-related barriers to attending international conferences,’ the report continues. ‘Even in highly ranked Rwanda, the Permanent Secretary of the Ministry of ICT reportedly estimated that the country had only about 10 AI engineers in November 2019.’

Few places offer a clearer glimpse of Africa’s AI-enabled future than Cotonou in Benin, the tiny tropical nation located on Africa’s Atlantic coast. That location is important because here, at one of the landing points of the global internet’s undersea cables, the Benin government has established Sèmè City, a digital hub that taps directly into the Africa Coast to Europe (ACE) cable system.

‘People come to Sèmè City seeking potential colleagues, training and backing,’ director Claude Borna told France 24. ‘Young people are often full of ideas but lack the framework and the backup.’

Borna raises an important point. Unlike, say, the US (38.1) or China (37.7), the median age in Africa is a young 19.5 years old – and that demographic of African youth is expected to double to 225 million by 2055. And while AI advances have thus far been driven predominately by the private sector, more and more African governments have been launching deliberate AI strategies for growth and governance. That’s empowering those young, tech-native innovators to create AI-based solutions for uniquely African problems.

When Benin closed its border with Nigeria during the COVID-19 crisis, Cotonou – which is situated a stone’s throw from that border – saw its imports drop to near-zero. But, as Borna told the Tony Elumelu Foundation, ‘although COVID-19 has been a devastating international crisis, for us at Sèmè City it has been a positive affirmation of how impactful our entrepreneur and youth-support mechanism has been. To date, entrepreneurs within our ecosystem have, in little to no time, successfully created and deployed solutions that include hand-washing systems, face masks certified for general public use, 3D-printed face shields, patient-tracking devices, medical professional support applications, food and essential needs delivery applications, educative content for children and much more’.

Or look to Tanzania’s cassava fields. Before opening its AI lab in Ghana, Google worked with farmers in Tanzania to develop an ML solution that helps to quickly identify and manage plant diseases. The team annotated thousands of cassava plant images, identifying and classifying diseases to build an AI-based mobile app based on Google’s open-source machine-learning library TensorFlow. ‘Once the model was trained to identify diseases, it was deployed in the app,’ TensorFlow’s Fred Alcober writes on Google’s the Keyword blog. ‘Farmers can wave their phone in front of a cassava leaf and if a plant had a disease, the app could identify it and give options on the best ways to manage it.’

In recent months, several African governments have established AI-specific agencies to drive the technology in local contexts. Nigeria has a newly minted National Agency for Research in Robotics and Artificial Intelligence. Uganda has an Industry 4.0 government task force. Kenya has its own 11-member blockchain and AI task force. South Africa’s Presidential Commission on the Fourth Industrial Revolution reported in late 2020 that AI is ‘a bedrock technology in the 4IR’, adding that ‘research and development, as well as implementation capabilities in Al, are thus critical and must be embedded within the state’.

Whether it’s COVID-19 interventions, agritech apps or smart-utility infrastructure, Africa’s AI innovators have long known what the continent’s challenges are, and how machine learning can help solve them. Now, with government interest and public-private infrastructure investment, they’re able to make those solutions happen.

By Mark van Dijk
Images: IITA, Reuters/Gallo Images