The Robot Apocalypse
By Judy Wajcman, Anthony Giddens Professor of Sociology, London School of Economics. She will open the European Forum Alpbach with an opening speech on August 16, 2017 .
Barely a day goes by without a news story, research study or business report declaring that the robot apocalypse is imminent. This is a rare moment where academics, entrepreneurs, journalists and politicians are united in fear, wonder and expectation.
No job will be safe, we are told. No section of society will be left unaffected. The robot apocalypse is coming and our entire existence – the way we think, the way we work, the way we live – will be transformed. How long do we have? Not long, apparently. It could be five years. It could be ten. It could be tomorrow. It could be today. Should we be worried? This question continues to divide opinion. Will the robot-run future be a place of conflict or cooperation? The jury is out.
An Ancient Anxiety
Our relationship to robots is like a case of schizophrenia. In one moment we see intelligent machines as miraculous timesaving tools that will make our lives easier, faster and more efficient, liberating us into a life of leisure and luxury. In another moment – and at times simultaneously – those very same robots are presented as an opposing force that will turn society upside down. At best, they will make the masses unemployed; at worst, they will make us all extinct.
A glance towards the past shows that this anxiety is not new. The fear – the fantasy – that our creations will rise up and overthrow us can be traced back to ancient Greece. Pandora was the first woman created by the Gods, under the orders of Zeus. She was endowed with many gifts – a mind, speech, beauty, strength – but she is ‘seductively devious’ and soon she has opened the jar from which all evil comes.
In modern science-fiction, different versions of the same story are told again and again. In 1818, Mary Shelley breathed new life into the tale with her gothic novel Frankenstein; or the Modern Prometheus. Now, the idea that our benevolent inventions will make us obsolete is the anchor of nearly every robot-related film, from The Matrix (2001) and iRobot (2004) to Her (2013) and Ex Machina (2015).
The robot apocalypse is no longer just a subject for science fiction, however. It’s front page news. Even serious newspapers like the Financial Times – normally immune to such sensationalism – write articles with phrases like “robots are coming for our jobs and we need to defend ourselves.” The Guardian recently asked: “are any human professions safe from automation?” Frenzy fills the air.
Again, however, even this media hysteria is not unique to our age. Over half a century ago, in 1965, Newsweek ran a cover story titled “The Challenge of Automation.” Back then, “automation” meant little more than an electronic thing that moved, like an escalator or a lift. But there was still cause for concern. “Businessmen love it. Workers fear it. The government frets and investigates and wonders what to do about it … Automation is wiping out about 35,000 jobs every week or 1.8 million per year”, wrote the reporter. Over the next fifty years, the economy certainly changed, but we saw an almost unparalleled period of growth.
Regardless of these historical precedents, today’s commentators nonetheless insist that this time it really – really – is different. This isn’t the latest incarnation of a recurring nightmare. This isn’t a dress rehearsal. This is the real thing. Yes, the masses that were thrown out of agriculture found jobs in factories; yes, there was the expansion of the service sector. But tomorrow’s robot apocalypse won’t be so simple. The future is here, and it’s scary.
Humans vs. Machines
When we speak of robots we typically mean two different things. There are physical robots: the machines that help manufacture our cars, the conveyor belts that circulate our luggage and the automated tellers or checkouts in banks and supermarkets. And there are virtual robots: the invisible algorithms that enable machines to perform tasks, whether that’s compiling a list of search results on Google, driving the cars that physical robots have created or recommending you a new book.
Both represent “threats” to traditional forms of employment, and already certain jobs have been automated out of existence. In the US steel industry, for instance, 400,000 people lost their jobs between 1962 and 2005, 75 percent of its work force. But its shipments did not decline.
Such figures are taken as proof that automation and the interests of workers are at odds with each other. It’s as if robots are to blame for displacing humans, as if they automate the jobs themselves through their own technical brilliance., while the inventors, managers and CEOs can only watch on helplessly.
But it’s easy to forget amid all the excitement over Artificial-Intelligence and machine learning that the replacement of workers is often the very reason why these machines are made in the first place. And a lot of the time, what makes them so effective has nothing to do with their technological ingenuity. Robots may be impressively fast, precise and consistent, but they are also cheap and don’t complain. For employers, in other words, they are the perfect workers.
In an interview last year, one of America’s most successful CEOs revealed his reasons for wanting to open a fully-automated restaurant. He didn’t want humans, he said, because they were “expensive.” Robots, by contrast, “are always polite, they always upsell, they never take a vacation, they never show up late, there’s never a slip-and-fall, or an age, sex, or race discrimination case.” Before his ambitions were foiled by the Senate, Puzder was President Trump’s pick for Labor secretary.
In this context, the reality of robots replacing workers isn’t so incredible. Indeed, the speed at which this takes place will say much more about the banality of the jobs we have humans perform as it will about our esteemed technological prowess. All the talk of making a machine that can finally “think like a human” masks the way in which we have already established machines as the standard in the working world. Again we see the schizophrenic mind-set: we want robots to think like humans and humans to work like machines.
According to this vision, humans are pitted against machines in a race to the bottom. Unable to compete, humans will be rendered obsolete, left to adjust to this robot-ruled world through increasingly precarious forms of employment. This is supposedly to the benefit of consumers who gain access to a wealth of services at the flick of a finger. But it is actually – as ever – the owners of the automated systems who reap the greatest reward from automation: profit, and without having to worry about the workers. The problem of bad working conditions is solved not by improving them but by removing humans.
A Universal Basic Income
Fears that humans could struggle to make a living in this technotopia are offset by Silicon Valley’s latest solution. A universal basic income – an old idea enjoying a new lease of life – will apparently ensure they have the means to survive.
This concept is growing in popularity across the world, but Silicon Valley’s enthusiastic support for it should make us wary. In practice, as the economist Thomas Piketty has written, a universal basic income ‘expresses a concept of social justice on the cheap.’ It fails to address the real issues involved in automation, such as the ways in which work is a meaningful activity for humans – not just as a source of income, but also of identity and dignity – while it leaves the structural inequalities of society largely intact. In practice, it may simply relax pressures on corporations to pay their workers a fair wage – a government subsidy for exploitation, in other words. As Piketty writes, ‘if we wish to live in a fair and just society we have to formulate more ambitious objectives.’
This rallying call applies to automation as well: we must consider more radical ideas for the future than simply the replacement of workers. For Silicon Valley, the future is a world where robots carry out our work and daily chores – drive our cars, deliver us pizza, order our shopping – while a few humans (invariably based in Silicon Valley) decide the direction of society and the rest stay at home, plug gaps in the system and do … what exactly? Probably watch TV, if the current habits of the unemployed are anything to go by. But Silicon Valley is not interested. It strives to streamline every activity imaginable without ever asking why.
If you want a clear vision of this future, I encourage you to watch Mark Zuckerberg’s latest promotional video for his ‘cutting-edge’ AI system, Jarvis. You wake up at home and Jarvis opens your curtains for you, tells you if your baby is awake, teaches them Chinese if they are, helps you get dressed, and makes you breakfast (it’s not clear how). It’ll even tell you what day it is. ‘It’s Saturday so you only have five meetings’, Jarvis tells Mr Zuckerberg. You’ve got to feel for him. Even in this world where everything is done for him, Zuckerberg still can’t keep a free weekend. Is this really the best the future can offer?
“We are being afflicted with a new disease of which some readers may not yet have heard the name, but of which they will hear a great deal in the years to come – namely, technological unemployment.” So wrote the economist John Maynard Keynes in 1930.
Keynes was writing amidst the ruins of the Great Depression in a society “suffering just now from a bad attack of economic pessimism.” But for him, this “new disease” was no “catastrophe” – it was a cause for optimism. Technological unemployment was a temporary affliction, a fleeting “phase of maladjustment”, he suggested, brought on by increases in efficiency outrunning “the pace at which we can find new uses for labour.”
After a brief period of turbulence, society would arrive at a “destination of economic bliss”: an enormously improved standard of life with working hours significantly reduced. “Three-hour shifts or a fifteen-hour week” await us, he predicted.
Keynes optimism was not unique. It was shared by many other intellectuals, like Bertrand Russell and even Oscar Wilde a few decades earlier. And who could blame them? If a machine(s) could do jobs just as well as humans, if not better, surely it was reasonable to assume that we would be able to work less?
“Thus,” Keynes prophesised, “for the first time since his creation man will be faced with his real, his permanent problem – how to use his freedom from pressing economic cares, how to occupy the leisure which science and compound interest will have won for him.”
Today, this “permanent problem” feels as far away as ever. Citizens of the wealthy West are on average four or five times better off than we were in 1930 – as Keynes predicted – but our average hours have fallen by only a fifth. Meanwhile ideas of the good life have all but disappeared from public discussion: our utopian horizon has shrunk to the size of the consumer self.
The belief that technological innovation could bring about improved social relations, reflected in the optimism of Keynes and Wilde in ‘The Soul of Man under Socialism’, has been replaced by this endless quest for automation, speed and efficiency. These values now present them as ends in themselves, the divine doctrine of progress. But as conceptions of cooperation between humans and machines go, they are very narrow ones. We should be able to imagine more.
Radical Thinking and Diversity
Genuine optimism for the future will require more than just robots, automation and streamlining. The speed of technical innovation is often conflated with inventiveness, but rapid technological progress can actually be conservative, maintaining or solidifying existing social arrangements. It brings forward a world in which everything changes so long as everything stays the same.
For radical change we require radical thinking. At its best, inventiveness should challenge our common-sense ways of doing things, interrogate the assumptions that permeate our political discourse and create new possibilities for the present. To borrow words from Walter Benjamin, it’s about striving for a world ‘in which things are freed from the drudgery of being useful.’
With the right policies and priorities, there is no reason why a robotics revolution cannot liberate uniquely human qualities like creativity, ideation, communication and empathy. We can build a world with less laborious toil and more enjoyable, creative jobs, not just the unfulfilling, precarious forms of employment that are currently on the rise.
Robotics will always be at the centre of our visions of the future. When we close our eyes to imagine the future, most of us will see a gadget or a machine doing something for us. We build our present and dream our future with and through technologies. The robot apocalypse may be imminent, but robots conquered our imaginations long ago.
To this extent, however, we leave engineering to the engineers at our peril. In my view, the homogeneity of Silicon Valley creators is a more dangerous threat to the future than any perceived robotic apocalypse. The small corner of society from which engineers are typically drawn is reflected in the narrow vision their inventions embody. I have frequently commented on the male-dominated culture of engineering, characterised by values of mastery, individualism and non-sensuality. Over the years, despite increasing attention to the subject of diversity, little progress has been made.
According to a recent analysis of 500 tech start-ups in the San Francisco Bay area with fewer than 100 employees, only 23 per cent of them are female. In the ten major tech companies, the numbers are often closer to 30%; but in technical roles, women occupy only 18.3%.
This is not an essentialist argument about innate female values. It is part of a more general proposition about opening up the processes of technological research and development to a wider range of societal groups and interests. Robots will not decide our future – we will. And so we need be more careful about who exactly that “we” is, and ensure it is as inclusive, democratic and diverse as possible.
We should all have a say in the future we want. The most prosperous future is one where machines and humans work together. Each one of us – individually and collectively – will always have something unique to contribute as creative collaborators with robots, regardless of how many robots there are. The challenge lies in building a world where we can.