Unemployment remains a pressing concern across Africa, where job opportunities are crucial for economic growth and stability. Amid the rising tide of artificial intelligence (AI) adoption, questions remain about its impact on employment dynamics, particularly in a region with unique socioeconomic challenges, and where approximately half a billion jobs need to be created by 2050.
In an exclusive interview, IC Intelligence sat down with Adio-Adet Dinika, a dedicated researcher at the University of Bremen and an intern at the Distributed AI Research Institute (DAIR), founded by Timnit Gebru. DAIR is “an interdisciplinary and globally distributed AI research institute rooted in the belief that AI is not inevitable, its harms are preventable, and when its production and deployment include diverse perspectives and deliberate processes it can be beneficial”.
Dinika sheds light on the often-overlooked realities of data workers and annotators who power AI technologies. His insights stem from his in-depth exploration of the impact of digitalisation on the future of work in sub-Saharan Africa, a focus of his thesis at Bremen University. Through this lens, Dinika delves into the challenges and opportunities that the region faces in terms of socio-economic development.
IC Intelligence: Many fear that AI will replace human jobs, particularly in regions like Africa where employment is crucial. Do you share this concern?
Adio-Adet Dinika: I would be naive to say that there won’t be job losses. There will definitely be job losses as a result of AI adoption. However, I don’t believe that these job losses will be confined to Africa alone. The effects will be felt as keenly as they will be felt in more developed countries.
Of course, what will happen is that the impact of job displacement will be felt more keenly in Africa due to poor social safety nets. But I also believe that Africa is well positioned, given its bulging youth population, unlike other continents with ageing populations. This can be viewed as an opportunity, as we are talking about a young, trainable workforce. If African governments engage in urgent programmes to develop an AI-ready labour force, then these job displacements or losses resulting from AI adoption will be less keenly felt.
Scandals erupted earlier this year over the controversial working conditions of Kenyan workers commissioned by Big Tech. How do you think this AI-ready labour force should be protected?
Without proper labour protections, AI adoption can exploit African workers, increasing economic inequality. Another problem posed by AI development or job digitalisation is the creation of a reserve labour force. The adoption of digital jobs has led to a high reserve labour force that can be exploited.
Therefore, ethical imperatives are needed to prevent the exploitation of African workers and to support displaced workers through transition periods. The responsibility mainly falls on African governments to swiftly engage in training programmes to prepare young people for these jobs.
Rwanda, for instance, stands out in this aspect, engaging in targeted training for its youth. This approach can be emulated by other countries to ensure people are trained and retrained to seize new job opportunities. Although AI may replace repetitive tasks, it also creates new jobs that didn’t exist years ago.
African youth need to capitalise on these opportunities. Furthermore, Africa’s youth demographic can provide a labour force not only on the continent but also beyond, given other countries’ ageing populations.
There is a fear that if businesses in Africa don’t adopt AI tools rapidly, they might struggle to compete globally. Where should we invest first when it comes to artificial intelligence in Africa to have the most impact?
Investment should go in capacity development because AI tools are only as good as the purpose they are created for. A quick move to adopt AI tools created from outside the continent will result in us having AI tools that are not fit for purpose. Most of these AI tools will need human babysitters because trusting AI tools, especially those developed for a particular need in a particular context, will result in serious ethical issues like the ones we mentioned before.
If African businesses quickly rush out there to pick up whatever AI tool is available, then they risk having tools which do not understand or respect African cultures and environmental context. Of course, I am not saying African businesses should not adopt AI tools developed by other developers. But before adopting, there is a serious need for evaluating these tools, because a blanket adoption of AI tools without simple common sense will result in serious problems.
The problems that African businesses are facing are not the same as those being faced by European or American businesses. In that sense, adopting an AI tool straight from Guangzhou or Beijing might not be the smartest business move for a company in Dakar, or in Lusaka.
AI models rely on reliable electricity, internet connections, and data centres. Should African governments prioritise infrastructure funding over AI tool development?
I wouldn’t say development of AI is a waste of money for governments. Yes, there is a need to develop that infrastructure because a stable infrastructure base can then allow us to exploit or to develop further AI tools.
Internet access is one serious problem that needs attention, because of the issue of having everyone aboard and not making AI tools or AI an elitist venture. So as long as there is no universal internet access in Africa, then it means that whatever AI tools are developed might not exactly be very reflective of the true nature of Africa, as they will be elitist ventures leaving out the poor and marginalised.
It’s essential to move forward on both fronts. Invest in developing locally relevant AI tools and focus on infrastructure to ensure diverse and equitable utilisation of these tools.
Tell us about your latest guidelines on how AI should be developed in Africa.
Technology is just a mirror of society. As I said before, it does not have any inherent magic, but what it does have is the priority of its creators, which is why I am emphasising the need for inclusion and participatory design. Because if we want to develop Africa, if we want to leapfrog other continents, we need to understand that AI is only as good as the environment it was created in. And by that I mean who created it, what data they used, what was the purpose of his creation.
So when we centre African needs, when we centre African ethical values into the creation of AI, then that is how African tools can be truly useful and beneficial to Africans and therefore, in that sense, help Africa solve some of its critical challenges.
Adio-Adet Dinika is a doctoral candidate and expert on digitalisation and the future of work.
