Robot-Proof Jobs 2: Transcript
Five years ago, Marketplace explored how machines, robots and software algorithms were increasingly entering the workforce in our series “Robots Ate My Job.” Now, we’re looking at what humans can do about it with a new journey to find robot-proof jobs.
David Brancaccio: Can I interest you in a big, beefy spider the size of a bicycle wheel that skitters across the floor?
Dave Rollinson: And then the legs, there’s a controller that basically figures out where the legs are now, to basically keep myself balanced. So I can drive it around, I can go up, I can go down.
David Brancaccio: Arachnophobes, be not a afraid. This is only a hexapod, with just six instead of eight legs. And relax, it’s a machine, a robot.
Dave Rollinson: Kind of modular building blocks is what we’re calling these. We want to get to the point where people can put these together as easily and intuitively as Lego.
David Brancaccio: A Pittsburgh company named HEBI Robotics has developed these interchangeable modules that serve as the shoulder, elbow or knee of any robot you’d care to construct, just add bits of tube springs, and foot pads, and voila, custom robot. Pounding the pavement with its clever little legs.
David Brancaccio: If technology is ascendant in the workforce, then being the person who develops the technology is a pretty safe place to be.
I’m David Brancaccio host of the Marketplace Morning Report. This is Robot-Proof Jobs, Part Two of a pop-up podcast series on a force that could be bigger than globalization when it comes to America’s prosperity. Machines, algorithms, artificial intelligence, software, robots: We know technology creates jobs but also fiercely competes for jobs. Automation is coming to the workplace, now more than ever. I’m asking what do we humans do to take advantage of this?
In the first episode, I started a road trip through the Midwest to meet people with jobs that, for now, are resistant to encroachments by technology, the folks robots are less likely to usurp. Now in podcast part two here, I want to spend some time with what may be the career-supreme for beating the robots at their own game: A job in the robot business itself. These tech careers are golden, so how is it possible there’s a such a shortage of humans to fill them?
Chapter 1: The Most Robot-proof Job of Them All
David Brancaccio: I first met Dave Rollinson here in Pittsburgh five years ago when I was doing a story on robots and jobs. Then, he was a graduate student at nearby Carnegie Mellon University’s Robotics Institute. He was working on snake robots for search and rescue and finding flaws in disgusting sewer pipes. Five years later, Dave has co-founded the company that made that big, crabby robot, we’re fooling with here. Old style robots are ungainly and can do damage if they misjudge the location of anything they touch. HEBI is commercializing the modular controllers that serve as the robots flexible joints.
Dave Rollinson: …Really don’t control position too carefully, just push down, and when it hits the table, it’ll just stop naturally, which is kind of a much more humanlike way of interacting with the world. But it’s very new to robotics so people are still trying to figure out how to kind of harness the power of being able to control force in a way that’s useful. But we’re getting there and we’re kind of on the leading edge of it.
David Brancaccio: With technology both assisting us on the job, and increasingly competing with us for work, one strategy to fend off the robots is to go Dave’s route. Become the creator of tech not its competitor. But doing a startup like this takes intestinal fortitude.
Dave Rollinson: You’re taking off a plane while you’re building the runway underneath it. It is that kind of level of stress.
David Brancaccio: Since we first met, a lot’s been going on with Dave, originally from the little town of Clifton Forge in southwest Virginia. He’s now Dr. Dave, PhD. And while he’s been giving birth to this company, he now has a spouse and a little human.
Dave Rollinson: A one and half year old, Amos.
David Brancaccio: Amos, congratulations, that’s wonderful.
Dave Rollinson: Thank you, he’s already better than the world’s best robot. It’s kind of amazing watching the trajectory there and how quickly they learn. His ability at a very early age to go to a bin and pick out exactly what toy he wants.
David Brancaccio: And, look, I knew it’s premature to ask, but I asked it anyway: What’s he going to do to help young Amos be ready for the uncertain job market of the future?
Dave Rollinson: Being a lifelong learner is going to be so important, just the ability to pick up new things quickly, the ability to have the passion to pick up new things quickly. Let him be self-directed as much as possible. If he wants to bang on the piano then we just plop him in front of the piano as long as he wants. If he wants to play with Legos, play with Legos.
David Brancaccio: When a job that doesn’t even exist today could be created and then eliminated by technology all before young Amos even enters the workforce, then education becomes a moving target. There’s a mis-match even now. Listen to Rollinson’s biggest worry on the professional front: finding people who have the right skills to work at his company.
Dave Rollinson: You’ve whittled your set down to probably like a handful of people in the world that can really do what it is that you’re trying to do. And there is no way around it being a slow, detailed process, but it’s probably our number one concern as we grow, is just finding the right people.
David Brancaccio: Across town, at Dave’s alma mater, we pick up this theme. David Bourne is principal systems scientist at CMU’s Robotics Institute. He’s taken me into one of his labs, a big open bay. It’s like that scene in every James Bond movie where you meet Q, the inventor, only it’s a graduate students doing the work here today.
David Bourne: So here’s Q’s assistant, this is Mabaran, who’s actually the real worker. He’s Mabaran, he’s awesome.
David Brancaccio: Mabaran Rajaraman is running a prototype robot that can recognize and manipulate big pieces, for instance girders on a construction site. Bourne says you can’t have more people with robotics jobs if there aren’t enough professors of robotics to teach them. Many potential teachers are being swallowed up by industry, like Uber, which famously hired away four professors and dozens of robot researchers from here to work on self driving cars.
David Bourne: Just to give you an example, in one of our programs we had 600 applications and there were 40 spots. So you think about that, that’s 560 students all of which are really qualified if you at their applications, you’re pretty impressed with the 560 that didn’t make it. That should give you pause, you know, there’s a lot of people that can’t do the field they want to do.
David Brancaccio: Across the country, another former student here is doing her part to train the roboticists of the future.
Anca Dragan: It was just what I was passionate about. I loved math and I did math competitions and the math olympiad. I was raised in a country where math is like our national sport, right?
David Brancaccio: Anca Dragan, originally from Romania, is a second student I met five years ago. Back in 2012, when she was still doing her graduate work, I teased her a little, do her parents call worrying if she’ll ever find meaningful employment with all that robot engineering she was studying? They weren’t worried then. Dragan’s landed on her feet, just fine. I’m speaking to her at her office in Berkeley, California, where she’s now a UC professor in the Electrical Engineering and Computer Sciences department. Her parents, happy finally?
Anca Dragan: They say, why did you not accept the Stanford offer?
David Brancaccio: But they must have been very proud of you.
Anca Dragan: They were, yes, absolutely.
David Brancaccio: Her research now centers on the relationship, call it the tango, between humans and robots, which can also be engineered to communicate their intentions non verbally.
Anca Dragan: We can’t help but read things into the actions that we see. We’re just wired to do that. We ascribe intentions, we ascribe capabilities or the lack of capabilities, we even ascribe some sort of, whether we want to or not, some sort of mental state. Like, oh this machine is excited or bored or whatever, right? And I think we need to get machines, robots to actually be aware of that to be able to take control over what they’re expressing to these people.
David Brancaccio: I love the story about her early inspiration, a book she discovered in high school.
Anca Dragan: I got my hands on Stuart Russell and Peter Norvig’s book on AI, and I read parts of it, whatever I could understand as a 12th grader.
David Brancaccio: AI, artificial intelligence, which prompted her to embark on a career of the future if ever there was one.
Anca Dragan: Now, I get to be a colleague of Stuart’s and he’s just a few offices away, so it’s really interesting to think of where I was in 12th grade and sort of the luck that I have now.
David Brancaccio: Well, “luck” in the sense of you can help make your own luck, but also the luck of being born in a country, Romania, with a culture that prioritizes math and science. Human math ability does tend to thwart the robot competition. My colleague, Marketplace Producer Katie Long has been doing a lot of the hard work on this road trip and she’s the keeper of the jobs and automation data we’re using from the McKinsey Global Institute.
Katie Long: Computer and Information Research Scientists, automation risk, 22 percent, pretty low. There are 22,000 in the economy. The pay is good, low six figures.
David Brancaccio: Professor Howie Choset is another renowned robot researcher at Carnegie Mellon. He’s also Chief Technology Officer of something called the ARM Institute, for Advanced Robotics in Manufacturing Institute. This is a national partnership with the feds, academia and industry set up to deal with the following stunning piece of news.
Howie Choset: We are the world’s leader in robotics research. We’re the world’s leader in so many things, but we’re not in industrial robotics. There is no major manufacturer of industrial robots in the United States.
David Brancaccio: Are you kidding me? Most factory robots are designed and made in foreign countries? I looked it up, the top five list of industrial robot companies goes like this: Japan, Japan, Germany, Japan, Switzerland. There’s an American company down at number ten. Choset and team have gotten serious money to turn this around.
Howie Choset: With any automation there’s always going to be some displacement. What we have to do is take care of that displacement. Retraining is a fact of life. So we in the ARM Institute, one of our functions is to provide workforce training experiences for small companies.
David Brancaccio: The top mission of the institute isn’t spreading the good word about robots. It’s listed as making U.S. jobs. The penny drops. Choset’s team feels the need to create something I would compare to a carbon offset. Instead of compensating for greenhouse gases, they want to compensate for jobs that robots might chew up. Technology creates work, but as we’ve been hearing it also replace workers, so the institute is forming a sideline to get people retraining into jobs that use robots.
Howie Choset: It’s just a fact of life. This automation is coming, so we should make that decision right now, do we want to be the leaders or the laggards?
David Brancaccio: I’m finding there is a paradox within the paradox here. Right now there is a shortage of people to do these technical jobs. The Manufacturing Institute in Washington predicts that by the year 2025, there will be two million unfilled jobs in the U.S. because there aren’t enough workers with the right skills. That’s two followed by six zeros, two million possible jobs unfilled within eight years.
Chapter 2: Getting Retraining Right
David Brancaccio: I said this road trip is about what to do about this automation coming into the workforce. How about this?
David Brancaccio: Retraining in action in Tecumseh, Michigan. I had gotten into a factory that makes structural parts for cars and trucks.
Steve Shafer: We make the GOR, the grill opening reinforcement or something, I think that’s what it’s called.
David Brancaccio: That’s the rugged metal nose, hidden behind all the decorative trim of a car or truck. A maintenance engineer here at Kirchhoff Automotive, Steve Shafer, is a man with burly arms, making burly parts for burly pick up trucks. The McKinsey analysis puts a job like that at what, Katie?
Katie Long: The person who runs the robots is an industrial engineer. Just 11 percent of that job is automatable, ironically. The average pay is alright, $81,000 a year, and there are 230,000 of these jobs in the economy.
David Brancaccio: This Kirchhoff factory has 670 people and is the largest employer this Michigan county. “We’re Hiring,” it says out front. The company works with state and local organizations to get people up to speed on the technology coming into the assembly line here. Steve himself trains colleagues. He equips people already on the payroll to be indispensable, even as more robots get hired.
Steve Shafer: We started out with certain types of robots and now we have four different style robots. So it’s up to my guys to learn what’s new, what’s different between this one and that one. How do I troubleshoot that and how do I fix this. You know, robots maybe taking people’s jobs, but you still have to have the person to work on that robot and work with that robot.
David Brancaccio: Because that’s higher level skills if you know how to fix it.
Steve Shafer: Yes, and most of that is training. It’s mechanical. It’s electrical. You know, robots are, people say, well they’re getting smarter every day, but actually they’re pretty dumb, they do exactly what people tell them to do.
David Brancaccio: There’s an outfit called Michigan Works Southeast!, a program that keeps track of what jobs are growing around here and gives out grants, state money, to get these skills into the hands of locals. Its director tells me the robot-proof jobs in high demand these days have titles like industrial maintenance mechanics, machine builders, and prototype technicians. Another dignitary from the local manufacturers association says it’s not feasible for companies to just hope people with the right skills will come knocking at the door, they’ve got to be trained. Programs like this one in Michigan are the gold standard for retraining now, they set up a feedback loop between businesses and trainers. It works best in real time for jobs that are needed now or pretty soon. Steve Shaffer knows the drill, given all his time in and around auto parts factories.
Steve Shafer: Quite some time. I’ve been here for 19 years and I was working at another place for 15 years in the same field. I started out as a mechanical, went into electrical, and then I went into technical. About 35-40 years.
David Brancaccio: So, you’ve done some retraining yourself.
Steve Shafer: Oh yes, everyday.
David Brancaccio: Never ends. Even outside of the world of manufacturing, companies are seeing the light on this kind of training from within. The global financial services firm Accenture says it brought in technology that cut 17,000 positions at the firm. But the story doesn’t end there. Accenture worked to retrain everyone to do new jobs that the company does need. The net result, Accenture claims, no layoffs from technology and jobs that are less, you know, boring.
As for other near-future options, wouldn’t it be nice if the roaring Silicon Valley job market could also be enjoyed in the land that jobs forgot, say the rustier parts of the rust belt or Appalachia? Strangely, it’s a politician from California’s Silicon Valley pushing just that idea. Here’s Ro Khanna, congressman and author of a book called Entrepreneurial Nation: Why Manufacturing is Still Key to America’s Future.
Representative Ro Khanna: I was in Appalachia, eastern Kentucky and folks there are being trained on Android software, iOS software for Apple, becoming graphic designers, web designers. So of course there is automation that is also displacing things, but what we have to figure out is how do we take advantage of technology, and how do communities take advantage of technology?
David Brancaccio: I have to ask, you represent a California district, what do you care about Kentucky or Appalachia?
Representative Ro Khanna: Well one, we need a greater skilled workforce in this country, and the companies that are in my district seek more people with the right skills, so workforce development is a national issue. And second, we can’t as a country find common ground if folks are left behind from the technology revolution. If we want to have any chance of healing this country, bringing folks together, we have to provide people with a concrete pathway to be productive in this new economy.
David Brancaccio: The term here, you used it, workforce development, that’s about education with a view to creating the skills needed for the jobs that are out there. Are we doing enough in this area of workforce development?
Representative Ro Khanna: We’re not and the problem is that some of the skills that we’re giving folks, or training folks, are ones that don’t lead to jobs and that’s why people are skeptical. Let me just return to this Paintsville example. Folks there were learning before the new program came, how to replace hard drives in computers, and that was not leading to any jobs because that’s actually something that could be automated. So there are two things that I think are critical. One is actually getting folks the skills in conjunction with talking to local industry that are gonna lead to jobs. And secondly, and this is equally important, for employers to commit to providing jobs and providing and investment in the training.
David Brancaccio: Getting retraining wrong ought to be a crime. Listen to the story of this man originally from the state of Maine. Robots ate his job and his retraining.
Dana Heath: My name is Dana Heath. I was a machinist for 28 years, from 1980 to 2008.
David Brancaccio: Heath was working up north for a company called Creative Machine.
Dana Heath: I was part of the shoe industry. We made aluminum molds for the soles of shoes.
David Brancaccio: Heath spent his days making these shoe molds, but even in his early years as a machinist, he considered shifting his career, so he started taking classes.
Dana Heath: In the 80’s I went back to school for computer, and went to night school for seven years, because I thought i was going to get out of the trade and get into something more lucrative. But by the time that seven years was up, I was far behind in the computers, and I had a bill of about $5,000 for the college. So I really couldn’t use what I learned, because I didn’t learn it quick enough. And I was getting raises on the machining side, so I said well, I’m making good money, I have a really good health program, I’ll stick with this.
David Brancaccio: He stuck with it until a new generation machines came into the factory, computer numerical control, CNC, units, where the machines do whatever a computer tells them. Machinists became computer operators.
Dana Heath: In 2008, I was let go. It was really through no fault of my own. It was because the production work had changed so much that I was kind of a dinosaur in that industry.
David Brancaccio: What to do? Heath had already tried retraining for a technical job and ended up with debt and less than cutting edge skills. If he was going to change his career again, he figured it’d have to be a dramatic switch.
Dana Heath: I went back to college to get a degree in social work because the calling that I felt was to do prison ministry. So in the end, everything worked for a better outcome really, but at that time I had no idea what I was going to do when I got let go.
David Brancaccio: I think the need to get this retraining stuff right is much more pressing now than it was when Dana Heath was making his career decisions. And that may have something to do with robot cars.
Chapter 3: The Next Jobs to Go Digital
David Brancaccio: Now, you may be saying, David, where’s the robot car in your robot-proof road trip? Behold:
David Brancaccio: What we have here is the robot version a Cadillac SRX developed at, yes, Carnegie Mellon University, said to be the birthplace of the autonomous vehicle, in this case a robot SUV. CMU Electrical and Computer Engineering Professor Raj Rajkumar.
Raj Rajkumar: We’ve been fortunate enough to be able to drive it safely in automated mode across multiple cities.
David Brancaccio: That’s automated mode 33 miles to Pittsburgh’s airport without incident. Another time across Washington, DC. This Caddy looks normal from the outside, fine Cadillac grille, no freaky electronic headdress or other weird protrusions. Sure, inside, there’s an explodey-looking red button in the center of the dashboard marked “Emergency Stop,” and the infotainment screen switches to radar to spot other cars, pedestrians, and, I assume, stray wildlife. There’s Marketplace Producer Katie Long’s low-rez image on the screen as she dances in front of the car. The stationary car.
Raj Rajkumar: See, that’s her.
David Brancaccio: The car showed up with a component missing on this day, so for this test, it’s hands firmly on the steering wheel in the time-honored tradition. But this encounter did get me thinking, maybe it’s the car itself that could be a road trip companion? Professor Rajkumar, serious engineer, will at least entertain my offbeat notion. He thinks the right metaphor might be Jeeves in the front, humans in the back.
Raj Rajkumar: They are sitting in the backseat, they’re getting some work done, taking a nap or doing something else while the chauffeur is driving. I actually think of Driving Miss Daisy, if you will. They become a companion or a colleague, if you you will. So you care about them, they care about you, and while they’re driving, you get to do what you please, do whatever you choose to do. So that’s the future that actually will result when the vehicle is driving itself, except that the chauffeur is not physical, it’s virtual.
David Brancaccio: The CMU team is hard at work on how autonomous cars will behave in complicated urban settings and dicey special case scenarios. All this is driven by two things. First, there’s a gold rush of capital chasing all this research right now, think Uber or think microchip-maker Intel spending $15 billion recently to buy a maker of robot car sensors. But Professor Rajkumar is driven by another imperative: 40,000 people died in car and truck crashes last year in the U.S., 1.3 million worldwide, most caused by people making mistakes behind the wheel.
Raj Rajkumar: Humans getting distracted, getting mad, getting sleepy and so on. If instead the vehicle is actually driving itself, most of those fatalities and or even more injuries and accidents are likely to go down, not necessarily to zero, but close enough to zero that society will be a big winner.
David Brancaccio: Robots could radically reduce the carnage, even if there’s other collateral damage in the labor market.
Raj Rajkumar: Across multiple states in the country, it turns out that driving is the biggest employment sector. So if vehicles can drive themselves, I guess there is some fear out there that jobs could be lost. So the good news to me is that since it will take upwards of ten years before a vehicle can itself completely without the aid of any human, those jobs are not going to be lost anytime in the near future.
David Brancaccio: OK, so ten years then. Just enough for bus driver Christina Summers.
Christina Summers: That’s all I ask, just give me ten. I’ll be OK with that.
David Brancaccio: We met Christina in episode one. When an algorithm took away her good job, she ended up driving a city bus which she hopes doesn’t get automated any time soon. So ten years, maybe. But it’s also true that Mercedes-Benz and one of it’s German supplier, Bosch, have just announced a collaboration to put a self-driving taxi on the road in just four years. This may be how the robots conquer the labor force: By car. Given the hope that autonomous cars will save tens of thousands of lives every year in America alone, these robots on wheels could become ubiquitous. It’s not robots vacuuming rugs at home or picking boxes in an Amazon warehouse that changes society. These cars and trucks remake our streets and our cities, they could remake the very nature of not just the labor market, but of the entire economy.
David Bourne: Driverless cars is the first big wave that’s going to crash over this society. I mean everyone’s going to face this.
David Brancaccio: That’s David Bourne whom we met earlier at Carnegie Mellon. He’s the first ever robotics professor there. And when it comes to tech and jobs, Dr. Bourne’s preferred frame is this one.
David Bourne: This is the real concern, it’s not what jobs robots are going to steal, it’s that people aren’t going to be ready to do the jobs that they need to do.
David Brancaccio: He argues that if we prepare, if, then things do move toward the better and away from worse, and not just for the people with the high-end degrees that his university gives out.
David Bourne: There’s going to be huge demand out there for people that know how to work with jobs doing the stuff that they’re capable of doing, it’s not going to be over their head. The jobs are going to be right for people. On the other hand, the nature of the work, the subdivision, what they’re used to working on, how they’re used to working, is going to change. And we need to get ready for that, that needs to be a priority.
David Brancaccio: In the third and final Robot-Proof Jobs episode, some radical policy ideas to help America get ready for the technology driven workplace. I have come across some revolutionary proposals that could mean rewiring the entire U.S. economy. In the next and final part of our podcast series: What if we owned all these robots?
Note: Transcript may contain typos or errors. Please cross-reference with the corresponding audio to ensure accuracy.
We’re here to help you navigate this changed world and economy.
Our mission at Marketplace is to raise the economic intelligence of the country. It’s a tough task, but it’s never been more important.
In the past year, we’ve seen record unemployment, stimulus bills, and reddit users influencing the stock market. Marketplace helps you understand it all, will fact-based, approachable, and unbiased reporting.
Generous support from listeners and readers is what powers our nonprofit news—and your donation today will help provide this essential service. For just $5/month, you can sustain independent journalism that keeps you and thousands of others informed.
Give today and get our limited edition tote.