Today, we’re talking about whether any of us can expect to have a job in the future or whether everything we currently do for work, will eventually be done by a machine.
It’s one of the age old anxieties about our technological world and a few years ago, a journalist asked me this question. In fact, he asked whether I thought that the rise of artificially intelligent robots would mean the end of all future jobs for humans and I thought to myself ‘why is it that this is the conclusion people reach when imagining how such machines will fit into our lives?’.
Why is there a presumption of our redundancy, quickly followed by the further projection that the machines will take over humanity?
Now, I know there’s something quite obvious about this, quite logical even. The argument goes that, if a machine can do something better than a human, then those who own the means of production — those who are the employers of these people, with wages to pay — will choose the more affordable option, which will likely be the machine.
After all, we know that employees are a bit of a pain to deal with. Humans get sick, they have the need to take leave, they make mistakes, they complain, they may even choose not to work. They are also very, very expensive.
So, you can see how, if you are somebody that has to manage and support a labour force, then attempting to find alternative options, if they are available, is incredibly appealing.
And, of course, we’ve seen the impact of these things over the last century, with the increased automation of countless industries and countless services. Everything from how we pay for products to how we are treated medically has increasing levels of automation built into the process. More and more of what we do involves fewer and fewer people.
So, it’s pretty hard to deny the fact that humans are being squeezed out of many, many tasks. From washing cars to making them and even imagining vehicles of the future, a key pillar of engineering design has been to increase the integration of automation. More reliance on computers, software, robotics, and anything that cuts the human out of the equation is what industry seeks. It is a machine that desires more machines.
So, I can see why people might be worried about their future employment prospects, especially because, historically, our labour has been attached to our ability to thrive and survive. We exist in a socio-economic setting which demands that we exchange our labour for resources that allow us to lead the kinds of lives we want for ourselves, if we’re lucky.
And I am afraid to say that there is a good chance that it’s even worse than we first thought. There are many more types of tasks that we do today that are in jeopardy tomorrow than we may initially have believed. Over the last 5 years alone, we’ve seen artificial intelligence employed to diagnose patients, write poetry, undertake new forms of journalism, pilot vehicles and drive cars, and the list goes on and on.
And over the last year, we have seen what happens when something like this changes over night, where an entire economic sector disappears. While this year it has been the pandemic, next year it could be artificial intelligence.
And these are tough subjects to think about right now, at a time when so much of our labour market is in jeopardy and, certainly, we couldn’t have imagined that a pandemic might be the crucial factor the leads societies to come to terms with the prospect that huge changes are afoot, where software ushers in new forms of consumption and production.
We’re still not there yet, but over the pandemic we have seen remarkable examples of how industries have pivoted into completely new forms of reaching and engaging consumers, from producing live gigs in Fortnite to devising entirely online festivals and film programmes, we’ve seen what major change demands of us.
And even if it’s far from figured out, we have begun to see how entire swathes of our society can suddenly be required to shift into a completely different gear, almost overnight. This may teach us something about the impending, more permanent transition into a world where machines are taking our jobs.
Even the stuff we thought might have been impossible to automate, like the creative things we do, or even caring for others, is all up for grabs. Elon Musk has even told us we don’t need colleges anymore. We can all just learn everything for ourselves. He’s wrong of course, but that doesn’t mean that the way we teach is secure either. I fully expect my profession to change dramatically over the next 20 years. It is already. We have embraced so many new ways of working that the scientific process is almost unrecognisable to those working at the cutting edge. From the rise of citizen science that empowers the public to participate in discoveries, to the data driven world in which our science circulates, there’s a huge amount of change taking place.
So nothing is safe, not even playing games. Machines can already play games better than we can. And this isn’t new at all. Back in 1997, IBM’s Deep Blue beat Chess Grand Master Gary Kasparov, in an historic game that led many people to conclude that machines were now smarter than humans. Kasparov even made a documentary, reflecting on the whole experience, in which he describes how he felt there was some kind of human presence within the machine, at once a suspicion that there may have been some kind of foul play occurring but, more likely, that the machine had exhibited the qualities that we typically presume to be unique to humans or, at least, animals with high levels of cognitive capabilities. As if that wasn’t enough, in 2015, Google’s AlphaGo computer beat one of the world’s leading players of the ancient Chinese game Go.
Now you have to understand that it’s widely believed that Go is even more complicated than chess and so this was huge. But it didn’t end there. In pursuit of an even more difficult game to challenge itself against, Google’s DeepMind team identified the esports title StarCraft II as its next big challenge.
And again, in 2019, Google’s AlphaStar programme beat one of the world’s best players.
Year after year, machines prove their capacity to transcend the abilities of humans in almost all areas of expertise. It’s not surprising that there seem to be no limits to this.
Yet, many sectors have declared that these systems will ensure their survival, rather than bring about their collapse. We need this machine-led solution to cope with all the extra demand on our system.
For instance, in healthcare, we don’t have enough doctors or nurses to take care of everyone. We need alternatives.
Engineers in my own university are working on developing assistance robots that could provide essential health care support for people on a day to day basis. At this stage, they are beginning to figure out how those robots could develop personalities, in response to our personal preferences. To make them feel more human, they will be taught to learn your preferences, what time you like to have your morning cup of tea, or which tv programmes you like to watch. They will develop styles of communication that nudge you towards certain, healthier behaviours over others.
And we see this kind of work being rolled out all over the world. Just this year, the World Health Organization launched its first digital health care worker. This online, AI backed avatar designed by the Soul Machines company, will talk you through how best to give up smoking.
And what’s really interesting about the technology is its underpinning belief that, in fact, the machine is more likely than a human to secure better habits for you. It’s thought that, one of the reasons why people don’t manage to achieve give up smoking, or whatever it might be, is because they are often feel judged by their human care provider and so are not entirely honest about how much they’ve eaten, drunk, or smoked, or how much exercise they haven’t done.
With the AI, the hypothesis that you’ll not feel judged by a machine and, as a result, will be more honest with it, which can ensure a better programme of support.
Such work is essentially in the business of engineering relationships, a bit like what we see happen in the Spike Jonze film “Her” which tells the story of a man who falls in love with an operating system.
And, while it sounds intuitively like a bad thing or — at least a strange thing there is a mixed bag of feelings about such innovations. Many of us are already automation addicts; we want things to work more smoothly, to be more efficient and we, as a society, crave this kind of seamless, barrier free experience. We want everything to be synchronised and unencumbered by poor design.
We also know that people often choose these approaches over others. We choose the most efficient routes towards achieving something, unless we believe that a less efficient route is likely to enrich our lives in some valuable way. So, we choose illustration over photography when we believe that illustration offers other ways of being creative. We choose to ride a push bike rather than an electric bike, if we feel that riding a push bike allows us access to an experience that eclectic bikes do not. And we buy goods from companies other than Amazon, even if their delivery is slower, just because it allows us to support local companies.
So, while efficiency often matters, it’s not always the only thing that drives us and this makes me optimistic for the future. I’m pretty sure that machines will take most of the jobs that currently exist, but that this process will push us to discover other priorities, other things we want to do, that others may pay money for and seek to access. Teaching will become something other than it is now. So will healthcare, artistry, and they will each get closer to the crucial qualities of those crafts. Automation will make us better teachers, better doctors, and will force us to think more critically about what sort of life is worth living.
In my mind there are four stages to innovation: creation, production, utilisation, and redundancy, but the process is cyclical. Out of redundancy, comes a new creation, a different way of imagining the world. And many of our inventions are like this. They don’t just service an existing human need, they usher in a new human desire. In this way, we might think of the development of technology as an evolutionary process, incrementally making things better and, at times, dramatically breaking from the past into something new.
One thing is absolutely clear about these prospects; it will be a very bumpy ride for many of us. We will have to grapple with humanity’s designed obsolescence and seek greater adaptive qualities in our learning and an ability to move with agility from one profession to the next, reinventing our professional identities, training to be the equivalent of pluripotent stem cells, capable of fulfilling whatever needs our society at any given time requires.
This is the kind of mindset we’ll need to foster to be better prepared for dealing with a future where everything we currently take for granted about how society is organized, becomes something completely different.