Opinionista Paul Zilungisele Tembe 4 October 2018

The robots among us – how should we manage them?

By their very nature, modern technologies are both disruptive and advantageous. One major complaint about technological breakthroughs in the Fourth Industrial Revolution is that they invade private privacy. Fourth Industrial Revolution systems like artificial intelligence are also accused of replacing human labour and the human mind. Are these accusations justified and what lessons can South Africa learn from countries like China and America on better ways of managing the Fourth Industrial Revolution and AI/IR?

From the start, South Africa has to leverage the unlimited opportunities that exist in artificial intelligence (AI). It can, and should, do this by moving from being a nation of passive consumers to providing solutions in conceptualising, designing, testing, and benchmarking AI products and technologies. This is so because in the Fourth Industrial Revolution, there is no one country that can cover all technologies that are needed for different sets of scenarios.

Linked to this, AI is important because it comes with national sovereignty. This is argued since it has to be asked: where is South African data stored and who owns that data in terms of intellectual property (IP)? This is a subject of paramount interest because there is no way that one can legislate on data that is stored somewhere other than within one’s own territorial space. In principle, I am of the idea that all data collected in a given national space should be under the jurisdiction of national government. However, the reality is that South Africa’s data is stored beyond its national borders and uses foreign licensed spectrums to various telecoms service providers.

AI is a natural evolution as we move from automation, information-based development to intelligence-based development that support, supplement, strengthen and expand human intelligence.

The Fourth Industrial Revolution is still an El-dorado territory where there can be win-win competition among different countries. The technological developmental spectrum is today wide enough for all to participate contrary to initial advances of the Second and Third Industrial Revolution.

Today, the world enjoys unprecedented technological leapfrogging opportunities that will lead to new ways of cottage education and cottage production systems powered by tools such as online crowd-funding, crowd-research, crowd-design in the form of 3D and Carbon printing technologies etc.

South Africa ought to develop its own AI-solutions based on its own social and economic conditions. Under the right conditions we would find AI everywhere in society, permeating each corner and every industry from industrial, township and rural economies. The world has never before experienced such a grand technological leap consisting a variety of potential initial nodes for a technology.

An area that may benefit from grand technological leap made possible by AI and other Fourth Industrial Revolution technologies is the overcrowded education system and start-up industries. 3D Printing and Carbon Printing technologies already provide solutions of distant learning, innovation and production systems. There are multitudes of technology forums in the web that are populated by like minded teachers that offer free tuition and advise on diverse future technologies.

South African students could easily tap into these forums of learning and innovation to learn, invent and produce solutions meant to solve local problems. In a simplified manner, the Elon Musks and Mark Shuttleworths are a product of exposure and opportunity to technology as it emerges at very young age.

AI will be a major driver for a wide variety of sectors like industrial upgrading, consumer markets, healthcare, education, smart cities, agriculture and defence systems. However, the design and roll-out of these systems in South Africa need to be inclusive of all three national economies: industrial, township and rural economies.

There are six factors that drive artificial strategy;

  • Robust Data Eco System should be in place to enable for AI to provide quality solutions at scale across education, health, agriculture and efficiency and safety in Smart Cities and Smart Mobility. Data ecosystem concerns matters of legislation, privacy versus public benefits of AI, intellectually property concerns that may lead to exclusion of greater mass of citizens where AI may be commoditised in a manner that serves private sector only. Data Eco System requires an adoption of AI across the value chain like start-ups, private sector, public sector undertakings and government entities to unlock the potential by creating a virtuous cycle of supply and demand. It furthers the creation of marketplace models focussing on data collection and aggregation, data annotation and deployable models. In simple terms launching of AI requires synergy between government and private entities if it is to deliver benefits the Fourth Industrial Revolution promises in its entirety.

  • Adoption of Traditional industries as decentralisation of the conveyer belt system of production model establishing adaptability of old industries into mainstream requirements of the Fourth Industrial Revolution.

  • Creation of new industries that arise out of AI and other Fourth Industrial Revolution technologies roll-out.

  • Specialised talent that responds to integrated skill sets required to fast track and develop AI and the rest of the Fourth Industrial Revolution technologies.

  • Education and training systems that are independent of centralised and sequential 2nd Industrial Revolution systems.

  • Ethical and legal consensus as in fast tracking legislation as not to cause bottlenecks in the roll-out of AI system and other Fourth Industrial Revolution technologies.

In AI China has advantages of government leadership, recruiting national resources, huge user base, mega project velocity, the world largest economy.

The US has advantages of first mover, top universities, diversity of researchers, venture capital echo system, openness and transparency, greater freedoms.

As a result both parties have experienced breakthroughs in big data intelligence, deep machine learning, brain-like computing, multimedia computing, human machine, hybrid intelligence, expert systems, even quantum intelligence computing.

South Africa fortunately possesses small pockets of advantages witnessed both in China and the US. The South African CSIR has robust technologies that if combined and acclimatised to broader South African conditions to include township and rural communities, the sky would be the limit for what the nation may achieve.

AI can be mobilised for practical purposes to solve real-life problems like our crime and safety issue. AI can be used to change our country’s status as a crime Mecca to a land of opportunity.

We have figures that show significant decreases in the crime rate with the application of video surveillance. It is evident from records from other countries that such technology can indeed turn South Africa into a safer place.

However, there are many issues involving privacy that we should be mindful of as huge amount of data are recorded some of which are irrelevant to criminal activities. Such data should be protected for people’s privacy. I am of the idea that public interest of fighting crime for everyone’s safety and protection of privacy is an issue that needs to be balanced and legislated into law.

South Africa is already a leader in the safety and security industry, all it needs is to take advantage of the emergent technologies to educate, innovate and produce solutions that hold the promise to unleash the nation’s developmental potential. DM

Paul Zilungisele Tembe is a Fellow at TMALI and Associate Professor of Zhejiang Normal University


Business Maverick

The R161-million SMS: FSCA slaps world’s second-largest fine on Jooste for Steinhoff insider trading

By Sasha Planting