By Tshilidzi Marwala – Not long ago, I asked one of my doctoral students a pertinent question about the validity of models developed by the vociferous personalities in the artificial intelligence (AI) community in South Africa as we battle the coronavirus pandemic.
AI is based on machines that are able to learn and make their own decision. This can be a useful tool in predicting complex situations, such as the trajectory of coronavirus, the impact on business, as well as any possible societal repercussions. Yet, it seems as though the models developed, which are incredibly diverse in their findings, have been wholly inaccurate.
South Africa began with three positive Covid-19 cases, then seven, then 13, then 16, and by mid-July, we had over 300 000 confirmed cases. As my student astutely explained, we are facing an unknown, which leaves us with many unanswered variables. Added to this, we are transplanting the global picture onto a local context. My student added an important reminder, while a robust AI coronavirus model has not been built, AI has been effective in other instances.
As the pandemic rages on, AI has been an important tool for diagnosis and prevention. For example, when the coronavirus unfolded in China in December, visuals on the use of technology were nothing short of something out of a science fiction movie.
In a video that went viral on social media in February, an older woman in China gazes upward at a drone addressing her. The translation, “Yes Auntie, this drone is speaking to you,” garners a slightly bemused smile from her, before the drone goes on to say, “You should not walk about without wearing a mask. You’d better go home, and don’t forget to wash your hands”. She heeds the warning and quickly rushes back inside.
Before the lockdown, in the Subways in Beijing, passengers were being screened for symptoms of the virus by AI and temperature scanners. Similarly, algorithms are being deployed to screen chest X-rays to differentiate between pneumonia, tuberculosis and the coronavirus. An AI programme has been developed that predicts with up to 80% accuracy, which coronavirus patients will develop severe respiratory disease. This can identify early on which patients will likely need hospital beds and which can be sent home for self-care.
There are, of course, limitations to current AI technology which is focused on reading pictorial scans such as ultrasounds, X-rays and CT scans. Yet, these have proven to be useful tools in managing treatment.
Of course, these are just instances in the healthcare sphere. As the pandemic threatens economies, including our own, research suggests that tapping into AI could be the answer to subverting this. Accenture research on the impact of AI in 12 developed economies shows that AI could double annual economic growth rates by 2035 and could increase labour productivity by up to 40%, increasing efficiency. Similarly, a report from PwC estimates that AI advances will increase global gross domestic product (GDP) by up to 14% by 2030, the equivalent of an additional $15.7-trillion contribution to the world’s economy.
While the importance of AI solutions is apparent, we cannot ignore many of the warranted fears that accompany AI. Currently, the AI landscape is uneven, with high concentrations of intellectual property and companies in the developed parts of the world. This presents a regulatory challenge because AI companies operate globally. Additionally, as technology advances at an exponential pace, policymakers and laws are struggling to keep up. Even more worryingly, those who control and own the most powerful AI systems in the future will increasingly have a great deal of control over the rest – this kind of power evokes concerns. In most industries, AI is set to take over many jobs, which will redefine the role of humans.
It is against this backdrop that on August 5, we will host the virtual AI Dialogue to address many of these fears and promote the need for the responsible use of AI in South Africa. The AI Dialogue is a multi-stakeholder event that aims to bring together South Africans in the AI ecosystem to try and find common ground in using AI as a driver for economic growth, social development and safety. This dialogue will involve policymakers, the technology industry, organised labour, start-ups and SMMEs, businesses, academia and civil society.
A responsible AI environment in South Africa will require an ecosystem that takes into consideration our unique context, constitution, legacy, history, culture, diversity and language. Women need to play a leading role in the AI ecosystem. Our labour market still favours men over women – why are we looking to adopt technology which could worsen our already wide gender digital gap?
As it stands, women in the country have lower digital literacy and less access to internet-based technologies than men. Young people also need to be at the centre of the AI ecosystem. The youth unemployment rate in South Africa reached an all-time high of 59% in the first quarter of 2020. Involving youth in the development of technology will promote the advancement of skills and creation of job opportunities.
Tech start-ups are leading the way in the development of technology and need to play a key role in the AI ecosystem. The private sector, academia and Government also needs to work together and invest in research and development, as well as innovation institutes; while at the same time practising responsible use of AI and safeguarding the rights and well-being of all South Africans. As we begin to ponder a post-corona world, this dialogue will be an important first step.
Tshilidzi Marwala, a professor and the vice-chancellor of the University of Johannesburg, deputises President Cyril Ramaphosa on the South African Presidential Commission on the Fourth Industrial Revolution