Everything began when someone, for the first time, crushed a nut with a stone when they got hungry. That is when the idea was born that tools and instruments existed that could serve mankind. The use of tools has continued to develop since then, with oxen and horses that have plowed our fields and carried our loads. Wind and water have ground our grain and crushed our iron ore. The breakthrough of the power of steam in the 1700s enabled us suddenly to transport people and objects over long distances. In the middle of the 1800s the industrial revolution arrived. Machines, conveyor belts, vending machines and later industrial robots have effectively replaced simple manual tasks. Machines are tireless when it comes to screwing on lids, folding cardboard boxes or soldering electronic components. No machine has ever complained to their boss, asked for a raise or taken a smoking break.
Technical achievements continue, as does the hunt for continuously- increasing productivity. But this has also changed our role in the workplace. We have stopped manufacturing products, and manufacture machines instead. As a result, many professions have disappeared. Between 2006 and 2011, about 10% of Swedish jobs disappeared because of automatization.1 Most of these were routine jobs in industry, at cash registers and administration, but a not-too-bold assumption is that we have only seen the beginning of this development.
A large portion of the work that we see as self-evident today will disappear in the future. With the encroachment of computers and smart algorithms we may see many other cognitive and thought-demanding professions automated.2 In the 2014 report “Every other job will be automated within 20 years” it is estimated that upwards of 53 percent of all professions in Sweden may be automated in the coming decades.3 That is equivalent to approximately 2.5 million jobs, and will result in Mayor cost savings for companies – just over 71 billion US dollars in lower payroll expenses when technology takes over.4
From an evolutionary perspective this is groundbreaking. Only thinking beings have been able to play chess, diagnose illnesses, drive cars and handle terrorists. But today we are developing machines that can manage these things better than we can. Many bankers, engineers, stockbrokers, economists, teachers, real-estate agents, pedagogues, lawyers and judges, for example, have long been protected from automatization, but can eventually rapidly reduce in numbers. What happens, for example, when an advanced algorithm can try more court cases in one day than a jury member can try in their entire lifetime? Or when a simple push of a button on a brain scanner can reveal fraud and lies with nearly 100 percent accuracy. In the not-too-distant future we will have such lie detectors in use. How many jury members and defense attorneys will disappear then?
Sometimes our lives seem better suited to robots, for machines that can be programmed. We work. We clean. We pick up the kids. | Danica Kragic Jensfelt, Professor in computer science, KTH Royal Institute of Technology
Even doctors are prey for automatization. IBM’s super computer Watson, which won the TV quiz show Jeopardy in 2011 against two reigning masters, is making a quick career in the world of medicine. In a recent test, Watson diagnosed a patient’s symptoms in ten minutes by running the patient’s information against 20 million existing patient journals. Doctors are astounded. Diagnosing patients at a much earlier stage makes it possible to take action that can be crucial to a patient’s survival, and even final healthcare costs. An AI like Watson has enormous advantages over a human doctor. First, it can store information about different illnesses, medications, research findings and medical statistics. For a human this is impossible. Second, it can compare this data with information about a patient’s entire health history, including family, neighbors and friends. Many believe that Watson may well become the world’s best diagnostician.
Just as Watson can ask questions and diagnose patients, we can see a similar development in sales, marketing and customer service. Machines can very well be successful sales reps that diagnose customers’ needs, help them with relevant information and finally assist in the actual purchasing moment. But they can also take over traditional work tasks previously assigned to market analysts. Martin Hoffman, CIO for Volkswagen, amongst others, contends this. He provides as an example their own virtual sales planners that have the capacity to analyze potential sales for more than 200 car models, with hundreds of accessories, in over 50 markets.5 Their accuracy is 90 percent. He maintains that this “bot” would be able to change the price dynamic on demand.
That more and more sales reps will be replaced by machines is something that is already happening. As early as 2013, two researchers from Oxford University predicted that many simpler sales professions would be automated in the future.6 Many signs point to their prognosis becoming a reality. According to the authors, the following sales professions are at greatest risk of eventually becoming automated (with the probability of automatization percentage in parentheses):
- Telephone sales reps (99 percent)
- Event sales (94 percent)
- Sales reps in shops (92 percent)
- Insurance agents (92 percent)
- Real-estate agents (86 percent)
- Sales reps in wholesale and manufacturing (85 percent)
- Advertising sales (54 percent)
- Sellers of technical products (25 percent)
Today, to mention a few examples, many telephone sales reps have been replaced with automated marketing, sales reps in shops with e-commerce, and advertising sales reps with programmed advertising.
Nevertheless, it is important to point out that automatization is not only about jobs disappearing. In many cases it involves a transition, where an old profession has lost its value but been replaced with a new role that requires new competencies. Therefore, we don’t need to face mass unemployment. The opposite may in fact be the case. In reality we will need more labor than before, which McKinsey7 and AI-expert Anna Felländer8 amongst others have predicted in their analyses.
By extension we can talk about two types of work that will remain: in part those jobs that create and manage systems of computers and machines, and in part those who work with things that machines and computers can’t manage. The latter includes value-generating and customer-oriented sales projects where we utilize man’s thus far superior abilities, such as emotional intelligence, feelings, understanding of complex connections, fantasy, creativity, new thinking and artistry. In short, it will be the different languages of feeling that will become the valuable human contribution to the business deals of the future. This and a substantial portion of problem-solving capability.
1 Fölster, S (2015, 17 April). Automatiseringen har tagit bort 450 000 jobb på fem år. Available: http://robotnyheter.se/2015/04/17/450-000-svenska-jobbhar-automatiserats-bort-pa-fem-ar/
2 Brynjolfsson, E. & McAfee, A. (2015). Den andra maskinåldern: arbete, utveckling och välstånd i en tid av briljant teknologi. Göteborg: Daidalos.
4 Lindström, K. (2017, 17 January). 2,1 miljoner svenska jobb kan automatiseras – med teknik som redan finns. Computer Sweden. Available: https://computersweden.idg.se/2.2683/1.673771/svenska-jobb-automatiseras165
5 Jerräng, M. (2016, 9 November). Så här kommer bottar att förändra företagen totalt. Och ja, de tar massor av jobb. Computer Sweden. Available: https://computersweden.idg.se/2.2683/1.669257/botar-forandrar-foretag-tar-jobb?
6 Frey, C. B. & M. A. Osborne (2013, 17 September). The Future Of Employment: How Susceptible Are Jobs To Computerisation? The Telegraph. Available: https://www.telegraph.co.uk/news/2017/09/27/jobs-risk-automation-according-oxford-university-one/
7 Lindström, K. (2017, 17 January). 2,1 miljoner svenska jobb kan automatiseras – med teknik som redan finns. Computer Sweden. Available: https://computersweden.idg.se/2.2683/1.673771/svenska-jobb-automatiseras168
8 Jungstedt, C. (2018, 8 July). Digitaliseringsekonomen: AI ger ingen jobbkrasch. Dagens Industri. Available: https://www.di.se/nyheter/digitaliseringsekonomen-ai-ger-ingen-jobbkrasch/