Rapid advances in technology, connectivity and telecommunications are conspiring to make Africa’s large, rapidly growing population a valuable asset for the automation revolution. It is imperative that Africa quickly develop agency in data and artificial intelligence and it will be lucrative for investors who support them by financing Africa’s telecom and data backbone.
Africa must urgently develop cogent digital strategy. This at first seems fanciful, or even superfluous, given the continent’s relative lack of more basic development. Indeed, there are myriad other challenges to which most would assign primacy. However, by setting their sights on participating in the ongoing fourth industrial revolution, developing nations in Africa may be able to chart a navigable course to rapidly raising living standards. With the window for pursuing labor led industrial development narrowing, Africa can’t afford to take a gradual approach towards rapidly matching prevailing technological standards. Several opportunities are open to Africa within the corridors of the coming age of hyperconnectivity and automation. Africa focused investors will be well served by a bold approach to the continent’s digital infrastructure.
One wonders what it must have been like to live through the bewildering gyrations of the industrial revolution. In more fanciful ruminations, one might imagine upheavals and turmoil as people try to cope with radical new technologies and abrupt discontinuation of professions, traditions and ways of life. However, the reality was far more mundane. Most people didn’t immediately realize that significant changes were underfoot. The steam engine, locomotive, electricity and even internal combustion were obscure novelties for decades before becoming sufficiently commercialized to become embedded in every day lives. It took longer still for these technologies to alter the production process enough to begin to eliminate certain jobs and alter modes of life. We tend to imagine the emergence of dominant technological innovations as the sudden bursting of a dam; the reality tends to be a steadily rising tide eventually flooding unsuspecting communities unprepared for the deluge of change.
A similar process is currently underway, with cloud computing, 5G telecommunications, machine learning and the internet of things gradually cresting our economic river banks. Today these technologies are the domain of specialist researchers and TED talks, but they are inexorably, if imperceptibly, seeping into all the seams and crevices of our everyday lives. The combination of these innovations portends a near future of widespread automation which will fundamentally reshape the global economy and are likely to reorder human societies. Sooner than most of us expect, the economic potency of highly efficient, self-directed machines will severely pressure wages and the aggregate demand for human labor. Already, machine-based intelligence has matched or eclipsed human performance in optical and speech recognition, medical diagnostics, war games and perhaps most noisily, chess. The supremacy may extend to legal drafting, investment management and architectural rendering – knowledge professions will not be exempt. Machines employing human like intelligence in narrow domains that can simultaneously execute an abundance of tasks with precise accuracy and don’t require food, breaks or weekends will create an extraordinary, and irresistible, economic asset for global producers and owners of capital.
The industrial commodity of this new technological epoch is data. Deep learning, the process by which an algorithm independently improves and refines itself by being exposed to vast quantities of targeted data, has sharply brought forward the timeline for when much economic activity will be governed by automated processes. The process is perhaps most easily understood when trying to comprehend optical recognition. Machines can be taught, for instance, to recognize a housecat – and distinguish it from other types of animals or even felines – most efficiently by exposing an object recognition algorithm to billions of individual images of cats and not-cats. Often the process also includes ‘reinforcement learning’, offering rewards for correct guesses by the algorithms and penalties for errors. This is far more accurate and expedient than trying to specify a list of rules for what visual inputs together comprise a cat to the exclusion of other similar objects. Such technology, called computer vision, is rapidly being applied to driverless cars, warehouse management technology, facial recognition and medical x-ray diagnostics.
Most fears in the popular imagination about the onset of the age of artificial intelligence conjure images of all-conquering and all-knowing sinister robots enslaving the world. In the near term, the real danger worrying the informed is the power these tools will imbue to those humans who will own the algorithms and pilot the applications driven by machine learning. Unseen levels of profits, wealth concentration and the attendant political – and perhaps, military – power that follow will enthrone a new order of global power. Ascendant powers such as China are working assiduously to understand and set the standards for artificially intelligent systems and the telecommunication infrastructure needed to bring it to markets and strategic theaters around the globe; traditional powers such as the US and European nations are just as keen to stake their ground. Just as multi-masted ships with lateen sails and the maxim gun helped the West ascend to the apex of global supremacy, these technological step changes will play an important role in the geopolitical destinies of tomorrow.
As such, Africa must not sit idly by as this ground shaking revolution takes shape. It’s geopolitical and economic future will be dictated by the level to which it has input in and ownership over these emerging technological standards. Moreover, the nascency of humanity’s understanding of these categories leaves open to Africa unusually rich opportunities to catch up. The African academy and the continent’s budding technology ecosystem, for perhaps the first time in a century, has an opening to participate in the determination of standard setting paradigms. Such defining technologies will have profound ramifications on the relationships between labor and capital, business and governments, private ownership and the commons, nation states and multinational enterprise – certainly Africa should develop its own opinions on the way our brave new world takes shape. If for no other reason, than the fact that the 3 billion plus Africans expected to be alive in the year 2100 will not be exempt from whatever new rules take shape in the age of AI. By that time, Nigeria is expected to be the world’s second most populous nation.
This demographic reality also injects a certain truism creating an opening for Africans to take part in these developments: with Africans set to comprise a large share of global population, AI applications will need African data inputs for the necessary accuracy and efficacy to be useful in a large portion of the global market. Companies such as China’s Transsion Holdings are already aware of this; it’s Tecno, Infinix and Itel mobile phone models dominate the African smartphone landscape – easily outselling Samsung and Apple AAPL – and in so doing are training speech recognition algorithms on Swahili, Wolof and Hausa while its computer vision systems are learning to identify the nuances of dark skin tones with data sets the vast majority of the world’s AI researchers currently ignore. Currently unnoticed edges in data sets such as this will be a critical competitive advantage in a near future where a much larger share of the markets in which driverless cars need to identify a human will be of darker complexions.
Banking, warehousing, organizing and transmitting the data that trains and develops ever more intelligent machine applications will be the critical path to success in this new landscape; data is the new crude oil. Enormous productivity gains will inure to those who master this artificial intelligence and thus maximize the output their societies can produce. The winners will be those that develop the best applications to process data, train machine systems and perform tasks on behalf of people. Just as in our current economic hierarchy, the rest of us will consume these applications and send surpluses and profits to those who do the hard work of value addition – storing, labelling and processing data. The first great fortunes in this brave new world will be made in the infrastructure that will effectively store, transmit and deliver this data and its applications.
This is already happening. The sharp increase in demand for data storage and remote connectivity during covid-19 forced lockdowns with billions across the planet working remotely has exploded valuations for so called “work from home” stocks. Cloud storage providers, digital productivity services, videoconferencing facilitators and online commerce will continue to be the big winners long after the novel coronavirus is brought to heel. The current pandemic has simply accelerated long developing trends in the evolution of the way we work together. None of these businesses are possible without the oft-unseen material infrastructure that makes our brave new digital world possible. Data centers, sub-sea cables, fiber optic connections, cell sites and radio spectrum form the unseen arterial network enabling the broad consumption of these innovations. These backbone system assets generally provide reliable yields increasingly favorable supply-demand dynamics.
In Africa, the investment case may be even clearer. During the pandemic, African business has suffered from lacking the sort of telecommunications connectivity familiar to the developed world. But herein lies opportunity. This has spurred an acute awareness amongst policymakers and business leaders alike on the urgency of moving more of the economy online. Payments and mobile money applications have led this move with a raft of mobile network operators applying for banking licenses and at least as many banks expanding their already robust digital offerings. Africa has actually led innovation in the digital financial space. Ghana has issued the world’s first digital finance policy as part of its Covid-19 response, accelerating the country’s move to a cashless society. Vodafone Group plc and Kenya’s Safaricom famously pioneered mobile money with M-Pesa, a strong precedent for the opportunities available to international investors in Africa’s digital advance. M-Pesa’s extraordinary 99% market share has driven consistent expansion of earnings and delivered enormous returns to its investors.
A few categories stand poised to offer similar opportunities to investors in Africa’s automation and connectivity related future:
- Data labelling: Labelling is the process of categorizing units of data – for instance, indicating that a photo an algorithm will use to train contains a cat. These are the nuts and bolts in the age of AI; the opportunities to collect, store and categorize data from within the continent and without will be legion. An attractive on-ramp for Africa’s workforce, labelling requires a fairly low bar for entry level skillsets and offers the opportunity to quickly up-tier human capital.
- Data Centers: Capacity utilization for Africa’s digital infrastructure is set to rise sharply due to the twin trends of explosive population growth and increased digital participation. Companies such as Facebook and Google are rushing to lay down fiber optic loops on the continent, while investors such as Nigeria’s Jamarkani family, Actis and Convergence Partners are aggressively expanding their footprint of carrier neutral data centers. These assets will essentially provide the real estate upon which applications will be built.
- Spectrum: The bands of radio spectrum capable of reliably and seamlessly transmitting the mushrooming reams of data that will support these budding industries are not unlimited. Licenses to operate these spectra will become an increasingly valuable asset to investors and companies that own them – and to the governments that regulate and sell them.
- Mobile network operators: With some of the lowest mobile penetration rates left in the world and the most novel subscribers available anywhere – 60 million new mobile subscribers will come of age in Nigeria over the next decade alone – MNOs still have an opportunity to develop trusted brands and relationships with the customers desired by automation and connectivity applications. Opportunities to transition legacy or struggling players into data focused competitors could offer rich rewards.
Africa can ill afford to be absent from this discussion and artificial intelligence applications will arrive on the continent much sooner than most assume. The continent fell behind global competitors in adopting new technologies once before; that had disastrous ramifications for the continent’s future. Another great game is taking shape. This emerging reality is an opportunity for both policy makers and investors alike: on a continent that is virgin terrain for many industries, the automation and connectivity boom is one of the few areas where Africa may find itself on more equal footing to the rest of the world – at the moment, no one has a clear lead on these sectors. That won’t be the case for long.