The fundamental need for social and physical distancing during Covid-19 has made it imperative that we hasten our adoption of the 4IR
Perhaps you may have noticed the barrage of black blocks posted on social media sites such as Facebook, Twitter and Instagram last week. In solidarity with the Black Lives Matter movement, which has seen a reemergence of protests despite the Covid-19 pandemic on the back of George Floyd's death, #blackoutTuesday trended. While the sea of black blocks was a powerful statement in itself, it was also an essential lesson in algorithms. Algorithms are set of instructions that allow a computer to operate. The Black Lives Matter hashtag has mostly been used as an educational tool, as a platform for activists and as a vehicle for change. Yet, as blackout Tuesday trended, many used the Black Lives Matter hashtag, which in effect clogged up the posts one could access. It was in effect, a silencing in itself.
The algorithms on social media sites dictate the order in which you see posts as you scroll through your feed, which is based on specific signals. Posts are prioritised and given more visibility based on a range of factors, including the time spent engaged with the poster, the number of followers and the kind of hashtags used.
Blackout Tuesday served as a lesson in the Fourth Industrial Revolution (4IR). While a powerful tool for messaging, for a movement, we must also appreciate how precarious Black Tuesday can be. Another social media platform, Tik Tok, which has been increasingly popular as countries across the world have had to go into lockdown, came under fire in the last week for effectively silencing #blacklivesmatter by not showing any posts under the hashtag. The company responded by saying it was a technological glitch that had impacted the hashtag views on a number of words.
This is perhaps, the most fitting instance of the 4IR in action as we honour Eric Molobi, whose birthday was this past Friday (5 June 2020). It was after matric when he began working as an electronic technician, and was the age of many of our students that he had his first encounter with racial discrimination that led to his politicisation. In a team of 18 people, Molobi was the only black technician, and often the foreman would order him to disappear for a day to elude being seen by a visiting government inspector.
It was only when he called at the local trade union offices that Molobi discovered it was illegal to employ black people in a skilled capacity. This spurred him to action. At 31, he was jailed on Robben Island for his political activities and imprisoned for six years. He used this time to study and obtained a BA degree through UNISA. It is here that his narrative changed. He was released in 1980 and was later employed on the Education Aid Programme of the South African Council of Churches under Reverend Beyers Naudé.
As unrest took hold of the 1980s, Eric Molobi played a pivotal role in the political landscape and was instrumental in forming the United Democratic Front in 1983. His role as a revolutionary in the sphere of education also evolved. He became the national coordinator of the National Education Crisis Committee (NECC), an alliance of high school and university student, youth and labour movements, which had been created as a response to the crisis in black schools. It was here that the vision for education policy after democracy materialised. In the post-apartheid landscape, he emphasised educational and community development.
Similarly, we are now watching our own narrative change. 4IR is changing all aspects of our lives. Advancements in artificial intelligence (AI), the internet of things, blockchain, biotechnology, material science and robotics are driving the 4IR. To understand this shift, it is imperative to trace back the previous three industrial revolutions. The first industrial revolution mechanised production using water and steam power, replacing cottage industries and manual labour. The second industrial revolution saw the introduction of electricity and mass production and changed the scale and speed of manufacturing significantly.
The third industrial revolution saw increasingly optimised and automated production lines through electronics and electricity. With electricity, each machine could be powered individually with its own electric motor. Similar to the industrial revolutions that have gone before it, the 4IR is poised to change every facet of society from the way we interact, to how our industries operate to the way we consume. It is fundamentally a paradigm shift. The 4IR is a meeting of the physical, digital and biological spheres through these technologies.
Some of our leaders, such as Eric Molobi, sought ways of improving themselves and searched for ways of studying under dire conditions while they served their sentences. You may ask why I draw parallels with this. The answer is that at a time when the nation has faced a lockdown, unlike any other experience we could have imagined, students have been required to continue learning online. And while this is the 4IR personified, we should pause and ask ourselves, what valuable lessons can we take from our "confinement", which, of course, is not comparable to incarceration. I have wondered what learnings were there between the four walls, especially for our political prisoners.
Would Eric Molobi have found himself as a hashtag and similarly silenced, deleted or an accidental victim of algorithms, which sometimes seek to hide inconvenient truths? While our world is dominated in all spheres by algorithms, I have just illustrated that these can be manipulated and are not always neutral. Yet, there are many instances where the technologies of the 4IR can add immense value to us. The lockdown, necessitated by the coronavirus pandemic, has brought this to the fore. Arguably, the fundamental need for social and physical distance has made it imperative that we hasten our adoption of the 4IR. As we've seen in recent months, perhaps in a more extreme case, is that the 4IR will permeate all aspects of our lives.
The call to action in the face of Floyd's death, for instance, has perhaps been the most amplified instance we have ever seen in the Black Lives Matter movement. It is worth noting that in our country, we have not seen a similar outcry for deaths under similar circumstances such as that of Collins Khosa, Petrus Miggels and Sibusiso Amos.
Eric Molobi's school of thought rings more true than ever now. If one examines the Floyd phenomenon, we have to ask ourselves what has triggered such a response. Why is this so? With many confined with little else but technology, there has been increased interaction with other users of social media, which in itself has served as a form of education. Many who have not been mobilised in the past have taken notice now. Of course, there is a dissonance between a social media activist and a real one. However, the uprising we see could be due to the new normal we see where technology, and particularly the technologies of the 4IR permeate into every aspect of our lives. But perhaps, this is now the most crucial time to take notice of the pitfalls of these technologies.
AI's function in simple terms is to mimic human thinking. Yet, this means it also mimics human bias. In 2018, Amazon had to scrap an AI recruiting tool that showed bias against women. The device reviewed resumés with the aim of mechanising the search for top talent, but by 2015, the company realised bias that has been built into the system, based on the dominance of men in the tech industry. The system taught itself that male candidates were preferable, and thus penalised resumés that included the word 'women' or 'female'. Similarly, face-recognition algorithms do not work very well for African faces.
Police in the United States use Idemia, which scan faces using algorithms, yet the results from the National Institute of Standards and Technology have indicated that two of Idemia's algorithms were pointedly more likely to confuse black women's faces than white women's faces, or black or white men. Where Idemia's algorithms erroneously matched the faces of white women at an accuracy of one in 10 000, it incorrectly matched black women's faces about once in 1 000 – so ten times more frequently. One of the reasons for this is because of the limitations of the African face libraries.
Another is the suboptimal data collection for African faces, which are different from Asian and European faces. A third reason is that we have not designed AI algorithms for face recognition from the African perspective, but rather from the European and Asian perspectives. This is not to say that we should eschew any mention of the 4IR; instead, there is a need to identify where the gaps lie and how we can address these challenges. The corona world we find ourselves in has served as somewhat of a yardstick for our preparedness for the 4IR. It has revealed where we can adapt, but it has also revealed the drawbacks of the 4IR. I would argue that this is perhaps one of the most crucial arguments for adapting the 4IR in our country and our own communities. We are hardly likely to ever be able to address these challenges if we only rely on the big players in China or the United States, for example.
In adapting to the 4IR, we are creating platforms, regulation and systems unique to our own circumstances. Of course, this calls for rational thinking leaders and citizens guided by the principle of the national or public good. In this era of bias entrenched in our technologies, with fake news rampant to mislead and a moral compass guided by social media, it is more imperative than ever that we have leaders of Eric Molobi's calibre. Far from being denied information in the 4IR, we are overwhelmed by it, and often it becomes difficult to discern what is true and what is not.
As an educationist, as someone who shaped policy after democracy, as someone who wore many hats as many are now urged, Eric Molobi's legacy serves as an important reminder that we must critically interrogate everything we are faced with to extract what is true and just.