There is an inconvenient truth that very few outside of academic circles are willing to discuss, but plenty have expressed concern about: artificial intelligence and automation are going to replace millions of jobs, and sooner than most people might think. This is an issue that goes beyond robots in factories or fully automated shipyards threatening the manual labour markets, it is extending into more specialist professions such as law and medicine as well, where algorithms are already being used to automate some of the simpler decision-making layers of the job. Ross Intelligence is an A.I. service for lawyers that takes on research work that would traditionally be performed by junior lawyers, while Babylon Health are working on an A.I. system that will diagnose health problems via an app in the same way a doctor would.
Other jobs including customer service, taxi drivers, executive assistants, shop cashiers, bartenders, construction workers, delivery drivers, street cleaners, accountants, bus and train drivers, and many many more are already seeing technology creep into their environment and take over much of the work that humans currently undertake. It’s a trend that is in its early throes but whose growth is exponential. The impact on the job market will be significant, but it goes much further than that — the concept of work itself is going to be challenged.
The industrial evolution
The last 250 years of Western civilization have been shaped by two major socio-economic revolutions: the industrial revolution and the technological one. The former cemented the foundations for modern capitalism by commoditising human labour and using mechanical innovation to increase efficiency. The latter is in its infancy, but has already made a huge impact in evolving the industrial blueprint even further to replace human labour with computer-driven systems. The problem is that the societal constructs that grew out of the capitalist workshops rely almost entirely on the creation of jobs. With technological automation we are hurtling headlong into a potential crisis, unless the system also evolves.
The Industrial Revolution, like the current Technological Revolution, was born out of innovation. Specifically, the invention of machines such as the power loom which allowed for the production of fabric at several times the speed at much lower cost than the traditional weavers could manage on their own. Initially, this early example of automation put many out of work, leading to protests, riots and violence. Over time, however, the scaleability of the machine-driven factory system allowed for overall production to increase which in turn fuelled demand from the export market that led to the creation of more jobs at home to meet that demand. The model so famously espoused by China, India and other emerging economies in the latter part of the 20th century was exactly the same model that led to first the United Kingdom and then the United States becoming the dominant economic forces in the world.
The model works as long as there are two key things to uphold it: firstly, there needs to be enough demand for the manufactured products to keep growth trending upwards, and secondly that demand needs to create enough jobs domestically to keep the majority of people employed. This delicate equilibrium in what Karl Marx called the Class Struggle is the key driver behind the boom years of the Victorian era in the United Kingdom and the first part of the 20th century in the United States.
The challenge, however, is in the nature of competition within a capitalist system that relies on free-market economics, as most do. The theory behind the free market is that everybody is able to compete, unfettered, with anybody else, giving everybody the opportunity to “go up in the world”. In reality, free market competition means success hinges on having the edge over your competitors, and that means focusing on anything that will allow you to reduce costs, increase production, and increase profits. Automation and technology, in this battle, are critical weapons. Replacing expensive, unreliable human labour with cost-effective, highly efficient machines is inevitable.
The algorithm economy
The impact of automation on the economies of advanced countries has been slowly expanding since around 1950, which through no small coincidence corresponds with the time when the modern computer was invented. The big industrial players of the 21st century are not the U.S. Steels or Fords of the world but the Apples, Googles, IBMs and Microsofts. Technology has firmly embedded itself at the heart of our modern economy and that trend is only going to continue.
The really interesting shift that these technology companies have brought about most recently is in the dynamic of work itself. Previously, work involved going to a particular workplace and using a machine of some description — be it a drill or a typewriter — to perform your duties. At the end of the work day you would down tools and go home. Nowadays, however, the tools you use for work are the same ones you use at home for leisure activities, and the internet supplies a permanent link between our own personal space, our work space and indeed every space. So while physical jobs are slowly being replaced by computers, effectively the tools of production are now in the hands of everybody.
It is no surprise, therefore, that some of the newest and most innovative businesses of the past few years have made full use of this fact. A great example is Uber, a pioneer of what is being described as the “gig economy” but which in many ways points to the future of the economy as a whole. Uber communicates with its entire ground-based workforce through an app on their phones. An algorithm determines which drivers are offered a particular pickup, algorithms determine staff performance and algorithms decide what everybody gets paid. Customers, meanwhile, also use an app on their phones and are also fully in the hands of algorithms to decide which driver they will get.
The gig economy’s main premise (or promise) is that workers are self-employed, free to work when they choose, with opportunities delivered to them on demand with the simple tap of an app. While the larger employers continue to automate their jobs, self-employment looks set to become an increasingly popular career choice. The technology behind Uber points to a future where algorithms replace sales and marketing, where services — and even products — are delivered to you via an app. We are already learning to trust algorithms where once we would have followed brands — the recommendation algorithms on sites like Amazon, Google search results, sponsored posts on Facebook, all using vast databases of our personal data to match relevant products to the people who might want them.
The future of commerce is, therefore, one where we will trust algorithms more than brands, and those algorithms will use the data we happily feed them to give us everything we need, wherever we are. The need for big businesses (except those that deal in data and algorithms) will fade away as small businesses and the self employed, connected to customers via algorithms and apps, will be much more commonplace. The trend is already underway, with self-employment having grown by 88% in the UK between 2001 and 2015 and showing no signs of slowing. In many ways this is a good thing: running one’s own business is repeatedly proven to be more rewarding than working for somebody else. But there will be short term pain to endure, as the economy is still based on the capitalist structures that rely on growth and job creation. Without the latter, more wealth flows to the CEOs while workers suffer from less job security. Without the former, competition becomes a liability as the market saturates and businesses dig their heels in. To get to a point where small businesses can truly thrive, some aspects of the socio-economic environment need to evolve. The technological tools also still have some way to go, but the opportunity is there.
For those thinking about making the leap into owning a small business, now is the best time to do so. Before everybody does.
Originally published in February 2017, but still relevant today.