Page 65 - UKZN Proceedings of the Conference Report

or not South Africa will be able to realise the economic potential of AI will depend on whether it will be able to put up the necessary resources for educational programmes, impartment of digital tools, and formulation of policies that will support innovation. However, there is also a potential problem with subverting already existing inequities, in that AI implementation varies by verticals and geographies. Smaller enterprises could become more digitalised but have fewer financial resources, compounding the issues that large companies face. Aghion (2023) warns that these issues will leave AI concentrated in a narrow range in South Africa, where only a few sectors and regions benefit while communities remain marginalised.
South Africa has its challenges, but on balance, AI is likely to benefit the economy and society in the country over the long term. As reported by Accenture (2017), the economic penetration of an AI bearer is expected to result in tremendous improvements in the growth of Western economies, increasing their growth rates by two times by 2035. While such increases might be more visible and pronounced in advanced economies, other countries such as South Africa may be able to harness AI-driven development if proper measures are taken to get rid of barriers. They present these approaches as focusing on human capital, enhancing the innovative aspect, and ensuring a conducive ecosystem for the deployment of AI systems. As Aghion (2023) observes, all these activities need to be complemented with others that address access to AI technology and its applications, especially for small firms and less advantaged segments of the population.
Movement towards some form of development or modernisation in South Africa will also require attention to the issue of the digital divide, which is a significant challenge in the adoption of AI in most regions of the country. It will be equally important for the public and private sectors to work together to promote equitable benefits of AI and the establishment of conditions and resources for steady expansion of the AI industry. In the event that South Africa succeeds in overcoming these challenges, AI could become an enabler of economic performance, jobs and societal progress.
To sum up, AI poses both threats and prospects for the workforce and economy of South Africa. It is true that AI will help create new types of jobs, boost productivity, and promote economic growth. However, the benefits could be outweighed by risks such as job losses, unethical practices in the hiring process and geographical inequalities with regard to use of AI. Therefore, the emphasis of policymakers must include development of the workforce, innovation and appropriate rules of engagement. The benefits of using AI should not only be noticed by a few members of the population.
All of these issues will be addressed, and therefore in future Africa will be in a position to use all types of intelligence effectively, including AI, with economic growth and development taking centre stage in development of the AI economy. Incorporation of AI in the employment sphere in South Africa can transform sectors, open new vistas, and boost growth, but it calls for a unified effort at all frontiers of society to harness its potential and minimise possible threats.
Governance and political stability
In the last couple of decades AI has developed exceedingly fast, such that machines can now carry out functions that used to require human intelligence. AI systems can not only analyse data and discern patterns, and even make decisions, but are also capable of surpassing human performance in skills such as learning, reasoning, perception and even problem solving. Unlike classical computing, which only functions with explicit commands provided, AI systems are built to handle large amounts of information, learn from the results and make predictions or decisions with remarkable accuracy. This technological advancement covers a wide variety of applications, such as natural language processing, image processing, decision- making and autonomous systems like self-driving cars. Considering the current growth in AI technology, its ability to change civil services has become a matter of much discourse. Peixoto (2013) stated that the history of technological advancement has left a number of issues concerning careful regulation of its use, including AI.
The issues arising from the use of AI in public administra- tion have been expounded on by several authors. Accord- ing to Peixoto (2013), many experts believe that public management can be enhanced by using AI, particularly because the latter improves efficiency and service provi- sion. In contrast, other scholars are of the opinion that it is too early to generalise about the impact of AI on pub- lic governance, as this greatly depends on how fast the technology will be availed in the public sphere, which will outdo the private sector. Over the years the govern- ment sector has implemented more strategies to adopt AI than the business sector. However, barriers to over- come – other than financial – include training, developing lawmakers, and citizen engagement (Straub et al. 2023). The World Bank Group (2016) and Peixoto and Fox (2016) claimed that previous efforts in the digitisation of gov- ernment services have not succeeded as intended. Kem- peneer and Heylen (2023) further explain that economic inequalities present another challenge to the adoption of technology by the government to incorporate AI. Insti- tutional capacities of the government are one of the de- terminants, as Van Noordt and Tangi (2023) demonstrate, of how well such technology can be integrated into public sector work.
Proceedings of the conference on Public innovation, develoPment and sustainability | 63