Anxiety follows automation – will smarter machines cause mass unemployment?

 |  Jun 30, 2018

Once upon a time, about as little as 12000 years ago, the only job humans had was to look for food – hunting and foraging and hustling for basic survival was the only challenge for us.

Then somebody discovered the power of putting some seeds into the group and sprinkling it with water. Before too long, food production became much easier. A single farmer in a few acres of land could easily produce more food than a dozen hunters scavenging through acres of lands and forests. This, by the way, was the onset of a “horrible” thing known as civilization.

Instead of foraging and hunting, this saved time was utilized to do other things, including pottery, urban planning, building homes, mining, and eventually, accounting, aviation, and medicine, all modern wonders without which the world we know would not exist.

One could say that this was the first time automation of an activity (agriculture) caused mass “unemployment”.

Just a few thousand years later, another group of people started harnessing the power of steam and create “terrible” things like locomotion. Instead of beating the horses to get to the nearest village, people could get to the other end of the continent merely reading newspapers. The employment of many little children for jobs such as chimney sweeping did reduce, but we found better methods. And eventually, these horse carriage workers and chimney sweepers moved on to do other, better things.

Automation, quite like any major revolution, is likely to cause chaos in the society. Like we saw during the agricultural or industrial revolution. It causes a lot of displacement, something we saw again when computing took birth. The people who were worried about computers doing spreadsheets in the 1970s couldn’t, even in their wildest dreams, have imagined a whole new industry created by computers only a few years later.

Back in the 90s, when our economy was opened up to the world, and technology exchange became faster, a lot of people were terrified that if computers came to India, they would suck up all the jobs.

What happened was the exact opposite. Computers actually created a whole new IT industry and our services sector bloomed – which significantly helped in our GDP growth and helped us fix our third world status. Our processes became faster which further resulted in more job creation.

Just take a moment to think – not before too long, there used to be long queues for ticket reservation which resulted in the wastage of a lot of time and energy just to get a ticket. Now, everything is online and can be accomplished in mere seconds.

Today, in 2018, there is a similar fear.

With the widespread adoption of AI-based technologies, automation has become a commonplace term. What has also become commonplace is the fear that these automated systems will eventually eliminate the need for any human intervention, and will result in widespread unemployment.

The fear, in all honesty, is not irrational. But, those who’re entirely sure that this will happen are quite frankly, missing out on a bigger picture. Let’s take you back to the pre-ATM times.

Before ATMs, there were human tellers deployed at various branches for the same purpose. When the idea of the ATM was brought forth, the world saw a lot of hues and cries. It was predicted that the introduction of Automated Teller Machines would result in a reduction of the number of human tellers needed at the banks. Oh, and that did happen. But what also happened, was that the cost-savings owing to the use of these ATMs enabled banks to establish all the more branches. This resulted in higher teller employment overall. In short, as automation evolved, it paved the way for new and improved job roles.

Eric Schmidt, the execute chairman of Alphabet – one of the biggest firms working with data and automation – claims that “A.I. will create more jobs that can’t be filled, not mass unemployment”.  Contrary to many who’ve warned us (and keep reminding us) against the potential large-scale unemployment, Eric believes that in order for things to function productively, humans will need to work alongside computers.  

The research arm of McKinsey & Company concludes that the near-term impact of automation will not be to eliminate jobs, but instead to redefine them. They suggest that the fear of automation should be seen as an opportunity for augmentation. Furthermore, there is a long way forward before we get to Artificial Super Intelligence (ASI), which isn’t something intimidating, either.

The example of ATMs we discussed earlier goes to show that despite automating the jobs of tellers, these machines didn’t really eliminate the jobs. Instead, they paved the way for improved operations, and therefore, created more jobs than ever. The positive effects of automation can be seen across a variety of domains and niches.

If we talk about the State Bank of India, they’ve automated their passbook printing. This allows their staff to attach more weight and importance to critical tasks like loan processing and currency conversion.

This same pattern can be observed across a variety of industries. Further, the jobs that were not even dreamt of a decade ago – data analysts, data scientists, information architects, user interaction specialists – have now sprung and are the more sought after job roles.

Although it is true that automation has taken over some blue-collar jobs in the past, the world today increasingly demands specialists in Artificial Intelligence and Robotics. Automation is undoubtedly to upgrade the white-collar jobs in ways that require better skill-set and roles. The jobs will in no way vanish but will need an improved outlook on things. Plus, there also might be several new jobs in the offing.

Automation means technology. Technology means imperfection.

However technologically advanced we are or might be in the future, these advancements are hardly flawless. Andrew NG – the founder of the Google Brain AI project and former lead of Baidu’s artificial intelligence team, notes that the tasks related to Artificial Intelligence are still relatively limited. He wrote in Harvard Business Review that despite Artificial Intelligence’s breadth of impact, a majority of its functions are still confined to situations where some input data is used to generate a quick response. In case of supervised learning, the type of software that allows this input-to-output production has an Achilles heel – it demands a considerable amount of data for training.

For instance, if we talk about the most popular use-case of Artificial Intelligence – the language translator, we’ll find that it requires a high level of human intervention. It needs humans to input the data and even create the program. To define the learning mechanism and pathway. These translation applications, again, are not flawless. They have a difficult time capturing the critical nuances of language. Even the previously lionized term “Big Data” is losing its charm and luster. Various examples of its shortcomings have proliferated, and the analytical analysis has become extremely standard.

With so much said and done, one thing is sure – the future does look exciting. What does it hold? That will be found out only in time, but here’s our take on what the future might hold.

How does the future look?

Are we heading towards an employment-scarce future, owing to automation?

“We are expecting 500 billion objects to become connected to the Internet, and this automation is going to hollow out middle and working-class jobs,” explained Jane Fraser, CEO of Citigroup’s Latin America business, while speaking at Fortune’s Most Powerful Women Summit in Laguna Niguel, Calif.  

However, this is just one side of the story – one face of the coin. Artificial intelligence and automation will also create new and previously unimagined jobs, too, in the future. Self-driving cars, for instance, may need drivers for emergency rides. There’ll definitely be an increase in the number of tech developers to cater the increasing need for chatbots for every industry. This will also result in the employment of more AI professionals who’ll give a human-like touch to these chatbots.

Just like websites and other applications, all the systems need continuous maintenance and updating. No matter how advanced or super-duper-efficient these artificial intelligent systems go on to become, some jobs will always be done better with a human touch. An analysis of the British workforce by Deloitte noted a profound shift over the past two decades towards “caring” jobs: the number of nursing assistants increased by 909 percent, teaching assistants by 580 percent and care workers by 168 percent. All these jobs can’t be automated and will require human hands for better results.

On one side, there are tech gurus ready to debate that machines and automated systems are prepared to take all the jobs. On the other hand, some economists and logicians claim that the AI revolution or any new technology, for that matter, will always create more jobs than they’ll destroy. The real answer, however, lies somewhere in the middle.

AI and automation will undoubtedly not lead to mass unemployment but is sure to shake up the foundation of many organizations that rely on outdated technologies. This will result in huge losses and then layoffs. The companies that, on the other hand, embrace this technological revolution, will disrupt the industry and speed up their workflow manifolds. This will further enable their employees to learn more skills and be much more efficient and productive than ever.




Getting paid was endless and many a times embarrassing series of follow-ups. With numberz it begins with a click and ends with a click! Sraboni Harlalka Partner at Wodehouse Capital

Join the 18,000+ companies that use numberz.

Learn More