Why I For One Welcome Our New Robot Underlings

by Andrew McAfee on April 9, 2013

Imagine a world where the robots did all the work.  They tend the crops, sew the clothes, cook the food, drive the trucks, and work on all the assembly lines in all the world’s factories.

In this world, everything would be a lot cheaper because labor costs would drop to zero. In fact, there’d be a startling abundance of stuff. And people would be freed up to do things other than work. We could use our time to explore, create, perform, craft, mingle, and so on because we wouldn’t have to work to produce the necessities or luxuries of life; the robots would be taking care of that.

I think most people, given the choice, would choose this hypothetical robot-powered world over the one we live in now. After all, it’s free of drudgery, toil, sweatshops, on-the-job injuries, stingy vacation policies, horrible bosses, and all of the other dysfunctions of today’s workplace.

Yet a lot of people seem unhappy with the steps being taken to bring us into this world. When I was a guest on the KQED show The Forum last month most of the questions and comments from listeners were about the evils of robotics, and how to keep them from taking jobs from people. And a survey just released by the Smart Manufacturing Coalition found that, as the Wall Street Journal put it, “When asked to describe the impact on the economy of modernizing factories with advanced technology and automation, nearly two-thirds of respondents said it either made no difference or actually hurt the economy.”

The reason for this hostility, of course, is that robots are perceived as taking jobs from workers today. And they are, just as the steam drill took John Henry’s job and countless other technologies have substituted for human labor.

In the past there have always been other jobs waiting for displaced workers and technological progress has not let led to mass unemployment. But today’s anti-robot sentiments spring from the fact that this time feels different. It feels like robots, AI, and other modern marvels are encroaching quickly and deeply into human territory, and reducing the need (and hence the demand) for human labor full stop.

As I’ve been arguing for a while here and elsewhere, I believe this is in fact the case. But I don’t resent the robots, or want to unplug or smash them. I want them to hurry up and finish taking over all the work so we can enter the hypothetical world described above. I want us humans to enjoy an abundance of both stuff and time, and the robots are the only way we’re going to get there.

I know there will be serious challenges along the way. In our book Race Against the Machine Erik Brynjolfsson and I included Voltaire’s quote that “work saves a man from three great evils: boredom, vice, and need.” We’ve got to figure out how people are going to pay for things (even when things are abundant and cheap, many of them are still likely to have prices) when their labor is not needed and hence has no value. And we’ve got to figure out what a meaningful and fulfilling life consists of when it’s not dominated by jobs and work as we currently think of them.

I don’t pretend for a minute that these are trivial matters. But they’re the inevitable challenges we’re going to face as we move from today into the bright future described at the top of this post. So instead of resenting or fearing the machines that are going to take us there, let’s start talking about how we want our future to look, and how to make that future a reality.

Resisting the robots is futile; they’re going to keep improving, and companies are going to keep buying and deploying them. And more fundamentally, resistance is deeply misguided because it’s exactly the same as resisting progress toward a post-drudgery, post-sweatshop world.

And isn’t that the world we want?

  • Anonymous

    I’m deeply puzzled by this. Almost all people who are really happy report their work as being a chief part of that. Research in your own book suggests that there is no built-in limit on automation that prevents it from taking over job functions that, if given the executive choice, people would prefer to do themselves. You are suggesting we give in to an enormous social experiment (a world without work) based on a generalization that few if any thinkers have ever endorsed (people can be happy without work). You seem to discount entirely the view, which many economists would support and to which your own book lends credence, that the circulation of capital will tend to draw in anything it can, even when that is detrimental for other parts of society.

    Your book makes one of the most cogent arguments around indicating why democracy needs to get a much more significant control over technological expansion, even putting the brakes on “innovation” that may ultimately be harmful to critical social institutions. Here, you say “resistance is futile,” a terribly antidemocratic sentiment. What if (a) resistance is not futile, and/or (b) automation and the elimination of labor is actually a really bad idea for human society and human flourishing? What if society would benefit from some kinds of work being automated, but we are also better off if humans do other jobs, *even if* machines can do them “better” or “more efficiently”? Is that an experiment we really should subject the whole world to and just see what happens?

  • Jed Harris

    I can see CloseToTheTruth’s point but don’t agree with it as stated. Open Source and Wikipedia contributors are good examples of how people can deal with the issue raised. They don’t get paid and don’t have jobs as such, but they find meaning and a degree of community in that work. Volunteers in many activities are similar.

    Perhaps it is useful to define “labor” as “Something you only do because you need the money.” Robots may free us from the need to labor, but we can still do meaningful work if we wish.

  • Jed Harris

    This is the clearest statement yet of McAfee’s new position, which has changed a lot since Race Against the Machine. In that book he and Brynjolfsson apparently expected most people to continue to get most of their income from their jobs. In this post it seems that McAfee now imagines a future in which most people don’t have jobs, but still have to get income to pay for things somehow. Many people today also rely on jobs for life meaning and structure, and social context. Without jobs most people will have to get those in other ways (as many do today).

    These two possible futures have very different policy implications. The “old future” where most people need jobs to earn money requires big efforts to upgrade human skills, increase job growth, encourage compliance on the part of workers, accommodate the demands of employers, and so forth.

    In contrast the “new future” described in this post requires big efforts to rework our institutions so people have adequate income and meaningful opportunities whether or not they are employed. This involves not just reshaping mechanisms, but also our norms, life expectations, and political frameworks.

    As McAfee discovered on Forum this reshaping isn’t going to be very easy, smooth or universally popular. I’m eager to hear his proposals for what the new future should look like, and how we can get there.

  • ChrisF

    Thanks for mentioning this, I too had noticed this shift in emphasis – I found the arguments for more effective education and training in RATM to be a little unconvincing, but I fully agree with the conclusions in this blogpost. Improving skill levels may delay the outcome, but they can’t change it. I agree that a fundamental reshaping will be necessary.

  • ChrisF

    Maybe it’s a lack of imagination on my part, but I don’t see how we can possibly prevent the emergence of these advanced AIs and robots. They’ll give such a huge advantage to those who own them. For argument’s sake, if the US were to ban research into such technologies, wouldn’t they emerge in other countries and have the same eventual effect on the global economy ? Is there an example of any area of technology that we’ve been able to successfully “repress” in this way, I can’t think of one ?

  • Jon Perry

    If you ask me, it’s more like “work CURSES man with three great evils: boredom, vice, and need.”

  • Curly

    Sounds nice. Unfortunately, some people are reaching that utopian post-work future ahead of schedule. Figuring out how to pay for things when you’re unemployable in America isn’t a challenge they’ll have to face eventually; it’s the daily lived experience of millions. And of course there isn’t the political will to grow the welfare state in the interim, to make the lives of the unemployed easier or less uncertain. The political class isn’t even aware of the transition taking place. Now that it’s being widely reported that the Social Security Disability Insurance has become a last resort for people whose unemployment benefits have run out, the question has become, naturally, how to kick the welfare cheats off the rolls. Things are bad and they’re going to get worse, and if you were the one teetering at the edge of the abyss, you might advocate going backward as well.

  • ChrisF

    Yes – that’s why I think this blog, and the recent media appearances by people such as Andrew and Martin Ford, are so important… a few years ago, I don’t think many people took the idea of technological unemployment very seriously. At the very least now, awareness seems to be building.

  • http://www.facebook.com/people/Alexander-Zhdanov/1232310317 Alexander Zhdanov

    “Something you do only because you need the money” is precisely how I, for myself, define *work*. Which is perhaps the major reason I react so acidly to the traditionalist slogan “get a job!” I eagerly look forward to a post-work world. Doing something you like, and would keep on doing for its own benefit, is hardly work. At least in my book. This is probably just semantics.

  • Jed Harris

    Agree with this and with ChrisF. I think we face a period of very rough contention over these issues. To resolve that we need big changes in the way we collectively think about and describe “unemployment”, “idleness”, “working” etc. We need changes in our ideas of what people deserve and how we should provide it. And so forth.

    As we try to make those changes there will be increasingly intense resistance from people who sincerely disagree, and from people and institutions that will lose out due to these changes. The losers will be those who’ve been able to gain power and wealth by leveraging “job creator” stories and the weak bargaining position of workers — and they won’t quietly give up their advantages.

    On the other hand large scale unemployment, resulting in large scale poverty is bad for the economy, and that is ultimately bad for nearly all businesses and nearly all people including most of the 1% As this reality sinks in there will be increasing efforts to find ways out of the box created by our current ways of thinking.

    So ultimately we’ll make the transition, but it could be a very rocky path.

    Unfortunately we don’t get have credible, concrete descriptions of futures where most people don’t labor and we still have all the goods and services we need. The attempts so far are either too vague or dismiss the most interesting questions with wishful thinking.

    We also lack realistic proposals for how to get from where we are to any good future. Before we can convince the larger public we need to have a story that could actually work.

  • Jed Harris

    Interestingly, you and CloseToTheTruth have exactly the opposite positions. He says “people who are really happy report their work as… a chief part of that.” You see something as work only if you don’t like it.

    I think this illustrates how hard we find it to get a grip on these issues. Even our language for describing them at the most basic level is extremely unstable and lacking in consensus.

    Of course in a way these are consistent, probably in many cases “people who are really unhappy report theiir work as… a chief part of that.” For most people these days their job just has a big influence on their life as a whole.

    I don’t think it is too hard to come up with terms that split out the various dimensions and conditions explicitly. But getting a broad consensus will be difficult.

  • http://www.facebook.com/robert.shaw.77985741 Robert Shaw

    If you haven’t already, read “The Lights in the Tunnel” which proposes a solution for getting income to people, even when they don’t have traditional jobs…

    It’s interesting that in RATM, the authors mentioned this book but then dismissed it as predicting the “end of work.” Seems like they are changing their tune and now starting to embrace the whole end of work idea…

  • http://www.facebook.com/tiago.tavares.142035 Tiago Rocha

    “I want them to hurry up and finish taking over all the work so we can enter the hypothetical world described above”

    This all that I hope.I hate my job, and it makes me sick, the techonoly may be a solution to free people like me from this…

    I am from Brazil and I see technology displacing people every month, the banks are one big example in my city.

    I ask to myself, how many time it will take.

  • Jake Myers

    An argument against the optimism…

    As one loses prospects for employment, he or she loses their own hopes for economic advancement. We are experiencing Rockefeller/Carnegie-era levels of income inequality in this country, and oddly enough, it’s occurring in the context of Friedman’s ‘Flat Earth’. Automation seems to be driving that trend in the wrong direction, yet we can’t have a sane dialogue in this country as it pertains to rational redistribution of wealth.

    And an argument for the optimism….

    While sitting atop a 600-year-old castle in India last year, I couldn’t help but imagine that while the great leader likely had access to all of the riches of his kingdom, he still had to do it every single day in 100 degree heat. Is that a life that you’d choose over yours today? Air conditioning is a scarcity we are actively eliminating in the second and third worlds, and perhaps technology and innovation may do the same for our other basic needs – around the world and for everyone.

  • http://www.krishworld.com/ krishnan

    Robert, interesting that you noticed this. Dr. Mcafee had till sometime back (I would say even during the time he visited Singularity University) had dismissed this idea (I remember him writing a post defending capitalism after his visit) on the need to rethink political framework and the idea of jobs in the age of abundance. Glad he is having a shift in the way he sees the issue.

  • phoneduty.com
  • Glenn Snead

    The Rise of the Machines is happening, but no one has an answer to the problem: “what happens when we have nothing to do?” Some examples are evident – France has a larger persistent unemployed population than the rest of the EU yet we don’t see a flowering of innovation, creativity, and production from this country. A few protests, a couple of general strikes, but for the most part we have an expectant and entitled population. Their new “rick tax” and the reaction of France’s few productive wealthy is evident. They haven’t created a post-wealth, or a post-work society. Merely an idle one.
    So if we eliminate work for the vast majority of our population, will our intelligence and involvement increase, or will we tune in to crapier TV and sink ever deeper into our couches? I think Isaac Asimov called it decades ago with his Three Law Robots. They eliminated all manual labor for the Spacers who eventually shunned all human contact, encouraged by their “safer than sorry” robots. Eventually the robots came up with a “Zero Law,” which was supposed to protect humanity itself but it caused inaction rather than action. Entire Spacer worlds were lost because the robots couldn’t decide how to best control the terraforming machines required to keep the planet habitable. So the Spacers died off while the Settlers, who limited automation and preferred to make decisions on their own, became the dominant human faction and settled the galaxy.
    So the question is: Are we Spacers desiring conformity, comfort, and a sense of security. Or are we Settlers who are willing to accept risk? The scenario painted by this blog’s author seems to put us firmly into the Spacer camp.

  • Anonymous

    Currently we are on the path towards seeming total abundance, but we all can rest assured we’ll never get there, because this path lies in the land of nearly total unemployment, and therefore through an unavoidable social crisis & reset of some kind.

    The hope that we as a society will figure out ways of distributing value based on a metric other than our labor contribution is probably just hope. We used labor, at least as an ideal, to determine value of each person, for eons, so any other kind of redistribution mechanism is going to be rejected as an extremely radical, not matter what it is – probably even in the face of major social crisis.

  • Geoff Roach

    It is interesting, and in some ways, I hope it comes within the next 10 – 20 years, but in other means, I hope it never comes. The way I see it, robots should only be doing the manual labor, jobs that are minimum wage jobs that hinder a lot of people. But what people should be offered is alternatives, what I mean by this is that we should offer free schooling to people to obtain better, wealthier jobs, jobs that really don’t work with robots (like doctors, actors, teachers, military, business, etc) what this will do is still give people work, but what it will also do is make it to where we have more people working in fields that are needed instead of forcing a lot of us to work at shitty 50-60 hour a week jobs making barely any money. So really this can work if done properly and I hope they do it right and give the people as a whole a better life.

  • Anonymous

    I think a deeper science-fiction look at humanity’s reaction would be Cordwainer Smith’s Instrumentality of Mankind. There, after some unspecified period of unrelenting safety, plenty, health, longevity, etc. society decided to deliberately reintroduce danger and uncertainty.

    Smith’s societal underpinnings were biological (the under-people, genetically engineered animals) rather than robotic, but the examination of humanity’s reaction is similar.

  • Anonymous

    Imagine a world where the robots did all the work. They tend the crops, sew the clothes, cook the food, drive the trucks, and work on all the assembly lines in all the world’s factories. In this world, everything would be a lot cheaper because labor costs would drop to zero. In fact, there’d be a startling abundance of stuff. And people would be freed up to do things other than work. We could use our time to explore, create, perform, craft, mingle, and so on because we wouldn’t have to work to produce the necessities or luxuries of life; the robots would be taking care of that.

    Oh, yeah? And how would we pay the rent? And there might be a startling abundance of cheap stuff, but there wouldn’t be a startling abundance of free stuff. And how would you buy even the cheapest of goods with no income?

  • Anonymous

    A counterpoint to your argument for optimism. That Indian king suffering in the 100-degree heat didn’t know any better. Oh, he might have at times wished it were cooler, but he never thought to himself, “I wish I had an air conditioning unit.”

    We can put ourselves in the place of that king today. Two centuries from now, assuming trends continue, people will recoil in horror at today’s barbaric medical practices. “They hacked open people’s chests? Just to transplant a heart? My god, what butchery!” Yet we don’t miss that Star Trek medicine. We think to ourselves, “80% chance of five-year survival after a heart transplant? My god, the wonders of modern medicine!”

    However, the problem with automation is not automation itself, but rather automation in the context of a capitalist system under which people just find work to earn money with which to buy goods and services.

  • stand up comedy

Previous post:

Next post: