Subscribe to John Mauldin's
FREE Publication:

Thoughts From the Frontline

Sign up for John’s free weekly letter and join 1 million of his closest friends.

We will never share your email with third parties

Outside the Box

AI, Robotics, and the Future of Jobs

August 19, 2014

This past week several reports came across my desk highlighting both the good news and the bad news about the future of automation and robotics. There are those who think that automation and robotics are going to be a massive destroyer of jobs and others who think that in general humans respond to shifts in employment opportunities by creating new opportunities.

As I’ve noted more than once, in the 1970s (as it seemed that our jobs were disappearing, never to return), the correct answer to the question, “Where will the jobs come from?” was “I don’t know, but they will.” That was more a faith-based statement than a fact-based one, but whole new categories of jobs did in fact get created in the ’80s and ’90s.

However, a new Wall Street Journal poll finds that three out of four Americans think the next generation will be worse off than this generation.

Barack Obama’s former chief economist Larry Summers began this chant of “secular stagnation.” It’s a pessimistic message, and it’s now being echoed by Federal Reserve Vice-Chair Stanley Fischer. He agrees with Summers that slow growth in “labor supply, capital investment, and productivity” is the new normal that’s “holding down growth.” Summers also believes that negative real interest rates aren’t negative enough. If Fischer and Fed chair Janet Yellen agree, central bank policy rates will never normalize in our lifetime. (National Review Online)

As the above-cited article asserts (and I agree), the term secular stagnation is a cover-up for the failure of Keynesian policies which, as my friends Larry Kudlow and Stephen Moore note, began in the Bush years and were doubled down on by the current administration.

Kareem Abdul-Jabbar, who is pursuing a career as a social commentator after dominating the NBA boards for so many years, tells us that Ferguson (which is on all the news channels all the time) is not just about systemic racism; it’s about class warfare and how America’s poor are held back.

Jared Dillian over at the Daily Dirtnap notes that the militarization of the nation’s police forces has been an issue for a number of people for a long time. Exactly why does some small community in Connecticut need grenade launchers? Seriously? But he does make a point about the news cycle and trading:

The key here is the press. Journalism is such a powerful force, a force for good or bad (often bad), but if you look at Radley Balko, here was a journalist who had a pet issue (which really made him a polemicist) and he kept writing about this issue over and over again, but nobody really cared. He had a small, but loyal following. Now he has a very large following, because this issue just blew up on national TV, and now everyone is interested in it, like, why does my police department have a tank? And so on. So it went from being a non-issue to a big issue – overnight.

So this is what happens: now that it is at the forefront of the consciousness of the press, any time they hear about an incident like this, they will report on it. And it will seem like police militarization is everywhere. Before it seemed like it was nowhere. The only difference is that now, people will be reporting on it! It’s not like this wasn’t an issue before Ferguson. It was a huge issue before Ferguson. But now, everyone is talking about it. People are interested in hearing about it. And journalists will report on things that people like to hear about.

But here is the great part. Our friend Radley Balko, the expert on police militarization, the guy who has studied it his whole adult life, wrote a book on it, is the leading authority, does he get to be the leading voice on police militarization? No. There are plenty of other opportunists around who just latch on to whatever is the new new thing and position themselves as the expert on it. I guess what I am trying to say is that the guy who is short the whole way up and is finally right at the top is actually worse off than the guy who is just right at the top. Think about the financial crisis. The cemeteries are full of the bodies of money managers who were smart and early and short all the way up. It really is about being in the right place at the right time.” (The Daily Dirtnap)

I think having a national conversation about the militarization of police is probably a good thing. We all support the police, but more than a few of us are becoming a little uncomfortable with the number of SWAT teams in our communities. A little balance here and there might be a useful thing.

But to bring us back to robotics and automation, Kareem and others do point out that the social fabric of this country (and of the entire developed world) is more fragile than we would like it to be. And while the country had 50-70+ years to adapt to increasing automation on farms from the 1870s onward, and survived that transition, the radical restructuring of what we think of as work that is going to happen in the next 20 years is going to be far more difficult. Especially when everything is on the news.

The report that is today’s OTB is from Pew Research and Elon University and runs to 67 pages. I have excerpted about six of those pages, which highlight some of the key takeaways from thought leaders among the 1,896 experts the authors consulted with, some of whom think robotics will be a huge plus and others who are deeply concerned about our social future.. (You can find the whole study at plus links in the first few pages of the report to other fascinating subjects on the future. Wonks take note.)

The vast majority of respondents to the 2014 Future of the Internet canvassing anticipate that robotics and artificial intelligence will permeate wide segments of daily life by 2025, with huge implications for a range of industries such as health care, transport and logistics, customer service, and home maintenance. But even as they are largely consistent in their predictions for the evolution of technology itself, they are deeply divided on how advances in AI and robotics will impact the economic and employment picture over the next decade.

The countries that are winners in the coming technological revolution will be those that help their citizens organize themselves to take advantage of the new technologies. Countries that try to “protect” jobs or certain groups will find themselves falling behind. This report highlights some of the areas where not just the US but other countries are failing. Especially in education, where we still use an 18th-century education model developed to produce factory workers for the British industrialists, putting students into rows and columns and expecting them to learn facts that will somehow help them cope with a technological revolution.

Finally, I note that our views on the future impact of robotics and automation have a tendency to take on a religious tone. While everyone can marshall their “facts,” the facts mostly get used to conjure up speculations about the future. This is not unlike some of the arguments I heard in seminary. Are you post-millennial or pre-millennial? Do you see a William Gibson post-apocalyptic world coming, or a bright Ian Banks future where technology is our servant and has freed us of the mundane drudgery of needing to work to survive ?

The transition we are engaged in is likely to be volatile, no matter what your religious (I mean scientific) opinion of it is. But one of the joys of my life – and I hope of yours –is that I get to live through it. This week’s Outside the Box is my hopefully helpful way of getting you think about these important issues.

This weekend was only partially aided by automation. We went out to my friend Monty Bennett’s ranch in East Texas, where he runs a wildlife game reserve. He has animals from all over the world. I think I saw more gemsbok at his ranch than I did when I was last on a South African safari. And to my great delight I saw a red Indian deer that had one of the most magnificent racks of antlers I’ve ever seen on any animal anywhere. And if the local redneck hunters knew about the size of the racks on his whitetail deer, he would have trouble keeping said two-leggeds from climbing the 8-foot fence that surrounds the property. (Please note, I grew up not far from Monty’s neck of the woods, where redneck was not a pejorative term.)

Have a great week, and remember that robots need jobs too.

Your wanting more automation in his life analyst,

John Mauldin, Editor
Outside the Box

Stay Ahead of the Latest Tech News and Investing Trends...

Click here to sign up for Patrick Cox’s free daily tech news digest.

Each day, you get the three tech news stories with the biggest potential impact.

AI, Robotics, and the Future of Jobs

By Aaron Smith and Janna Anderson

Key Findings

The vast majority of respondents to the 2014 Future of the Internet canvassing anticipate that robotics and artificial intelligence will permeate wide segments of daily life by 2025, with huge implications for a range of industries such as health care, transport and logistics, customer service, and home maintenance. But even as they are largely…

Discuss This


We welcome your comments. Please comply with our Community Rules.


Craig Cheatum

Yesterday, 5:44 p.m.

I would suggest that software can replace 100,000’s of white-collar jobs without any new technology.  We can all think of dozens of jobs in healthcare, insurance and financial services that can be eliminated by a few keystrokes of programming.  The second reason is that there are too many operators in these sectors. There are some 1,300 healthcare plans for example.  How many does the economy really need to optimize outcomes?

Craig Cheatum

Nick Proferes

Aug. 20, 8:07 p.m.

“The view from those who expect AI and robotics to have a positive or neutral impact on jobs by 2025”

My reaction on reading most of the comments in the article has been: Where have these people been the past 40 years?
As someone who worked at the coalface in manufacturing for 20 years in design, production engineering, and management, I’ve watched the evolution of automation and robotics first hand.  I then spent the next 19 teaching engineering and design at a tertiary level, a lot of it in the workplace, and training the next generation of managers, designers, and change agents.  There are a lot of valid points in the article but a lot more Pollyanna-ish thinking.  So, some comments on some of the comments:

The Optimists:

“Vint Cerf – “Someone has to make and service all these advanced devices”

Indeed but in my experience in broad based manufacturing and related service industries, automation results in a NET loss of jobs.  Productivity improvement, an absolute necessity to maintain competitiveness is key.  High volume manufacturing of most of the items we buy, autos, appliances, hand tools, etc. typically has direct labour content as only about 4 to 8 percent of product cost.  The rest is material cost and overhead, including all the non-value-adding activities such as back office paperwork, material handling, and inspection which are now being automated as well.

“Michael Kende - In general, every wave of automation and computerization has increased productivity without depressing employment, and there is no reason to think the same will not be true this time.”

As above, I just can’t agree with this, the jobs “created” by automation, say manufacture of the machinery, tooling, service of that machinery, etc. most often move offshore.  Germany, for example, has benefited from it but at the expense of so many more jobs in its customer countries who benefited only in improved productivity from German engineered and manufactured machinery and tooling. 

“Argument #2: Advances in technology create new jobs and industries even as they displace some of the older ones”

Indeed they do, but its overall numbers we need to be concerned with.  Consider all the jobs lost on a car body building line stacked against those relative few created in designing, manufacturing, programming, and servicing the robotics.  The loss is a net one and goes further than just this one example.

“Amy Webb - But we will still need folks to do packaging, assembly, sales, and outreach.”

Actually we don’t, most of those jobs, if not all, have been automated, at least in some industries.

“Argument #3: There are certain jobs that only humans have the capacity to do”

“David Hughes - I don’t think all ‘self-driving’ cars will be totally unmanned. The human’s ability to detect unexpected circumstances, and take action overriding automatic driving will be needed as long and individually owned ‘cars’ are on the road.”

We will see, but the technology for self piloting aircraft which can take off, fly to a destination, land, park and shut off their engines has been around since the ‘60’s and is in common use today, if only in drones at this stage.  Trams and trains are run without operators in many places around the world.  Though the technology is there for cars, even if and when it’s implemented, I don’t see how that application can have much impact on work and employment.  We rarely hire someone to drive us to work, though cabs are a factor.

“Pamela Rutledge - An app can dial Mom’s number and even send flowers, but an app can’t do that most human of all things: emotionally connect with her.”

True, but this discussion is about jobs, and who is going to pay for that and how much, to help pay the mortgage and put food on the table?  How much unpaid “human” service such as raising kids, doing housework, looking after aging or disabled people is there already is society?  And this is going to increase as the population ages.

“Argument #4: The technology will not advance enough in the next decade to substantially impact the job market

Another group of experts feels that the impact on employment is likely to be minimal for the simple reason that 10 years is too short a timeframe for automation to move substantially beyond the factory floor.”

Where have they been the past 10, 20, 30 years?  It’s been happening and at an increasing rate of change.  No one knows what sort of technological breakthrough will happen next week let alone in the next ten years, but even existing technology is changing the nature of work and the workforce at an increasing rate, as product life cycles shorten, productivity needs to and does improve, and costs come down but so does overall employment in that sector.

“Argument #5: Our social, legal, and regulatory structures will minimize the impact on employment”

It will have to; otherwise face huge social disruption, even anarchy.  But how?  Increased tariff barriers? Increased wages?  Higher taxes on those in the highest paying jobs?  All of these solutions push jobs offshore or otherwise disrupt trade to the detriment of the country’s economy.  Who will do the lowest paying jobs?  And will they earn enough for a “living wage”?


“Mark Nall - The social consequence is that good-paying jobs will be increasingly scarce.”

This already seems to be a well-established trend in many areas of employment.

“Justin Reich - There will be a labour market in the service sector for non-routine tasks that can be performed interchangeably by just about anyone—and these will not pay a living wage”

I expect they will just HAVE to.  The US already has 47m living below the poverty line and the social disruption is becoming more and more evident.  Eventually something has to give.

“Point of agreement: the educational system is doing a poor job of preparing the next generation of workers”

I agree with this entirely having spent half my career in education (and trying to change that system based in the 50’s and 60’s).  So much now relies on “on the job” or “in-house” training to really prepare the workforce and even senior management to deal with the rapid changes in work practices and job knowledge and skills needed.  This points to failures in the public education system.

“Point of agreement: the concept of “work” may change significantly in the coming decade”

Peter Drucker argued this more than 20 years ago.

“Hal Varian - The work week has fallen from 70 hours a week to about 37 hours now, and I expect that it will continue to fall. This is a good thing.”

Indeed, job sharing where a 37 hour a week job is effectively shared between two people who only want to work part time is becoming more common. BUT if the working week is to be effectively shortened for all, say to a 25 hour one, society at large must be prepared to pay a price for this in terms of higher costs of goods and services or jobs will be outsourced to places where people are still happy to work a 70 hour week in order to support their lifestyle.  The “world” will simply not allow this to happen unchallenged.

“Peter and Trudy Johnson-Lenz - Many things need to be done to care for, teach, feed, and heal others that are difficult to monetize. If technologies replace people in some jobs and roles, what kinds of social support or safety nets will make it possible for them to contribute to the common good through other means?”

Unpaid work, I suspect, as is now the case.  A lot of “traditional” jobs are now undertaken by volunteers, and as the population ages, this may well continue to accelerate.  Does this take “paying” jobs away from those looking for work? I suspect it does.

“Tim Bray, “It seems inevitable to me that the proportion of the population that needs to engage in traditional full-time employment, in order to keep us fed, supplied, healthy, and safe, will decrease. I hope this leads to a humane restructuring of the general social contract around employment.”

It will have to as I suggested above otherwise how will society, families, keep up with the rising costs of housing, food, transport, etc?  But its a high hurdle to get over as evidenced about all the opposition to “socialized medicine” of the ACA which if more properly implemented would have done an even better job of reducing this important cost for families.

“Possibility #3: We will see a return to uniquely “human” forms of production

Another group of experts anticipates that pushback against expanding automation will lead to a revolution in small-scale, artisanal, and handmade modes of production.”

Already happening in both manufacturing and even service industry such a restaurants who are concentrating more and more on locally sourced and fresh produce, but increases costs and will it really replace traditional manufacturing and service at a sufficient rate?  Will we see chain restaurants and high-volume car plants go out of business?

This is recognized in the closing paragraph, but I wonder about this conclusion:

“In the long run this trend will actually push toward the re-localization and re-humanization of the economy, with the 19th- and 20th-century economies of scale exploited where they make sense (cheap, identical, disposable goods), and human-oriented techniques (both older and newer) increasingly accounting for goods and services that are valuable, customized, or long-lasting.”

As goods and services do, and always have, go through a life cycle starting with low volume, niche based provision though increases in volume to consolidation and eventually becoming a commodity.  Morgan are still hand crafting cars from wooden frames and hand worked metal, and still making a profit doing so, but the real volumes, sales, profits, and job numbers are still in repetitive, automated manufacturing of cars.

So, I tend to side with the pessimists and my conclusions:

Where will the new jobs come from? 

New businesses, start-ups which typically provide the greatest share, some 70-80%, of new employment, while more traditional, high volume manufacturing tends to lose jobs as productivity improves through automation and other means.
Typically new start-ups can’t afford high capital cost equipment and are more reliant on direct labour
But as growth from niche markets to higher volumes occurs, that will change this as it becomes easier to justify capital expense.

New technologies will open new opportunities, as has new techniques for oil and gas extraction, alternative energy, through wind solar and water, logistics as inter-continent trade continues to increase, moving goods around the world is becoming a more and more important element of product cost (which is not true of a lot of services, like, for example, call centres). 

AI and automation can be applied to all of these endeavours, but will, over time, change them and result in a net loss of jobs in these enterprises as new opportunities arise.



Aug. 20, 10:17 a.m.

It is hard to believe that a dramatic increase in productivity doesn’t imply dramatic opportunity: shorter work days, fewer dangerous jobs, less boredom, etc.

The big problem is deciding how the fruits of improving productivity are distributed in the population, and that primarily involves finding ways to redistribute income without destroying the incentive to be useful on the part of either the recipient or the donor. We are already trying to deal with that problem, and unfortunately have learned a lot of ways that don’t work. But we have more years to experiment before failure might lead to killing the goose that’s laying the eggs.

Giovanni Isaksen

Aug. 20, 6:42 a.m.

The quote from Henry Ford is the key: “[H]e does no good to his business if his own people can’t afford to buy the car.” Since 70% of our economy is based on consumer spending, automation can only succeed to the point that it provides the majority of people to have work that allows them to survive and have hope for a better future.

David Stahl

Aug. 20, 5:40 a.m.

It seems obvious that automation, AI and robotics can be used to replace lower skill jobs more easily than “knowledge worker” jobs.  I think this has already contributed to the unemployment and income distribution problems in the U.S.  This, along with offshoring and massive immigration at the low end.

Carl linden

Aug. 20, 4:11 a.m.

Actually, both views are probably valid. Long term, technology will probably make things significantly “better”, in the sense of how better is commonly used. In the short term, one to three decades, it will probably be quite disruptive on the job market, in particular, those who have completed their education without a set of skills enabling them to compete effectively in today’s and tomorrow’s world. Eventually, hopefully, the education system will get “fixed” and get caught up, but it will not be of much help to most of those in their 20’s to 50’s who are ill prepared for the new challenges. So, the transition to the new world that is really technology centric will be painful and disruptive, with potentially severe consequences to the social fabric. Heck, it will probably be a decade at best before our politicians even get serious about recognizing what is going on as an issue and begin meaningful discussion on sensible actions.

Gordon Davis Jr

Aug. 20, 1:43 a.m.

In 10 years, the benefits of automation will probably outweigh the risks to employment.  However, in 20 or 30 years robots will exceed human capability in a wide variety of endeavors.  The notion that human cognitive ability will allow adaptation and thereby ultimate benefits from the advance of robotics is flawed.  In a generation, robots will be better at virtually everything.  The other issue that seldom is mentioned in these discussions is the effect of population growth on our collective futures.  There are arguably too many people in the world right now.  The growing world population is about to run into a revolution in robotics which will require fewer individuals to provide all the goods and services demanded.  Many challenges await.  The millennials will get to sort it all out.

Aug. 19, 5:03 p.m.

United States jobs can be divided into two categories, those that are exportable to cheap-labor countries and those that are currently non-exportable.
Those that are exportable will be among the last to be automated.  Robots will have difficulty competing with low wage Chinese.  There is little financial incentive to automate anything when the workers only make 1/5 of what American workers make.  Chinese jobs will be safe for a long time.

The robot invasion will target the remaining non-exportable jobs currently held by Americans living and working in the United States.  There is great incentive to eliminate high labor cost to make products cheaper.

Automation may take forms other than human-shaped robotics.  We are already seeing driver-less vehicles where the whole car is the robot.
Good middle class jobs will continue to disappear.

Aug. 19, 4:30 p.m.

There are solid points made on both sides of this issue.  However, one point that is not being made enough is demographics.  The stark reality is that many industrialized nations are aging and many millions are set to retire during this time period (2015-2025).  As these large numbers of people retire, someone…or something…is going to have to fill the void.  In service, robots will take the non-customer-facing jobs but won’t go much further.  In manufacturing, customers will have a choice: either buy more stuff from parts of the world that have more people than jobs, or make more domestically with the use of automation.  Which is more politically palatable—Made in America (by robots) or more imports from elsehwere?  I would put my money on the former.

The stark reality is that world population growth is slowing while expectations for standard of living are rising for billions.  Someone…or something…is going to have to do the work in order to maintain those standard of living expectations.  There will still be plenty of work for real humans to do—customer-facing jobs where social skills are important. 

I don’t discount the pain this will cause in the short-run: it is real.  Unlike previous waves, this next wave (which will target some white collar jobs) will hurt the people of wealthy nations more than poorer nations.  But like free trade, the net result will be positive.  The key is whether the populace will accept it as well as free trade has become accepted in the last 25 years.  As in free trade, countries that don’t embrace it risk being slowly left behind in the long run.