Getting Ready for Artificial Intelligence in Sub-Saharan Africa
By Ernest Amoabeng Ortsin*
Image: Pixabay
Artificial Intelligence (AI) is the latest catchphrase in the technology space. Virtually everyone is talking about it. What seemed to be sheer fantasy in Hollywood movies a few decades ago is now here with us in reality. Machines are now learning, talking, seeing, sensing, listening, walking and running. In short, machines are now acquiring or mimicking a lot of human attributes and capabilities.
In the technology world there is already talk about a 5th Industrial Revolution (5IR) where humans are expected to share the same office space with machines as work colleagues. For people with doomsday beliefs, this seems very much like an Armageddon-like dystopia. They ask, how can machines or, more specifically robots, work with human beings in the same office? Can something possibly go wrong?
Well, while you ponder those questions, I want to bring in the African dimension and the fact that there is a clarion call on the continent to take advantage of AI. African governments are being called upon to invest in AI because it has the potential of helping the continent address some of its socioeconomic challenges in health, education, agriculture etc. But, the question is: what sort of AI does Africa need? Given the high rates of unemployment on the continent, do Africans need machines that would compete with them over job opportunities?
With these questions in mind, some experts have been advising the continent to focus on areas such as health and agriculture where there are some low-hanging fruits to be plucked. In Sub-Saharan Africa (SSA), for example, the World Health Organization (WHO) in a 2013 Report indicated that there are about 5,000 patients to one doctor. In general, the Report revealed that there was an estimated deficit of 1.8 million health workers and this figure was projected to increase to 4.3 million by 2035. Clearly, this is a big challenge for the SSA region.
In terms of agriculture, there are a number of trans-boundary pests and diseases such as locusts, armyworm, fruit flies, banana diseases, cassava diseases, and wheat rusts that cause huge losses to crops and pastures and threaten the livelihood of vulnerable farming populations in the rural areas. On top of this, climate change is also affecting crop yields, leading to serious food security challenges.
Against this background, some experts have proposed that AI capabilities can be deployed to help the health sector through record-keeping, triage and diagnoses of ailments at hospitals and clinics where there are no doctors. It has further been proposed that AI can be used in the diagnoses of crop diseases and also provide information on weather patterns as a mitigating measure on the impact of climate change.
The simple question then is: how ready are Sub-Saharan African governments for AI? More specifically, how are they addressing the following fundamental prerequisites for AI?
Strategic National Policies on AI: It is important to mention that the benefits of AI do not accrue in a vacuum. There is always the need for carefully-tailored national policies. And these policies must spell out the plans and strategies for the deployment of AI in the relevant sectors of the economy. At the present time, many SSA countries do not have such policies and this does not show readiness for AI. It is therefore recommended that SSA governments should, as soon as practicable, expedite action on such policies.
Data Governance on AI: Data are the oxygen of AI. Without data, there is very little that can be achieved with AI. Unfortunately, in many SSA countries, there are no effective mechanisms that govern data in terms of its generation, processing, storage, retrieval, usage rights etc. This is quite disenabling. It is therefore recommended that SSA governments put in place mechanisms that will facilitate the generation and usage of quality data.
AI Investment Funds: AI is expensive. It requires a lot of research and development (R&D) investments. For example, according to an MIT Sloan Management Review Report (conducted in 2020 in collaboration with BCG GAMMA) an estimated US$37.5 billion and US$50 billion were spent on AI research, globally, in 2019 and 2020, respectively. Out of this amount, only about 10 percent has been recouped by the companies that made the investments. Obviously, this is a big challenge for startups in SSA countries that can barely pay for their operating costs. This means that SSA governments should be ready to shoulder some of the costs associated with AI R&D. Internationally too, there is need for some investment funds on AI targeting SSA, however little.
AI Ecosystem: AI is not a standalone system. Rather, it is a melting pot. It thrives on the synergies of different players including machine learning experts, data scientists, web and mobile application developers, systems analysts, policy analysts, state regulators, investors, startups, tech-hubs, etc. In essence, there is need for SSA governments to facilitate the building of AI ecosystems that are capable of delivering state-of-the-art solutions.
In conclusion, it must be reiterated once again that AI holds a lot of promise for the future of Africa, particularly the SSA region. But, before that can become a reality, the region needs to plan for it. Governments in the region must be ready to support the private sector to nurture the right ecosystems. Lessons from the COVID-19 pandemic are clear: technology is the future. Africa must get ready!
* Ernest Amoabeng Ortsin is an Africa-based researcher with a growing interest in Artificial Intelligence policy research. He studied Political Science at the University of Ghana. He was a participant in the Leading with AI Lab and is a co-founding member of Leading with AI.