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.
Introduction
David Brancaccio: I always think that artists are the ones who see the future first.
Dawn Hancock: You don’t really get a chance to design a transportation system very often so it’s pretty pretty fantastic for us.
David Brancaccio: In Chicago, the bicycle sharing system is called Divvy, as in, let’s divvy up this fleet of bikes. Their livery, light blue with the double-V chevron motif, was designed by Dawn Hancock and her team at Firebelly Design. I stopped in at Firebelly to check out a robot-proof job for people whose gifts may not be in math or science, but instead in art and design. The portfolio here includes the look Fireberry designed for a coffee place called Half Wit.
Dawn Hancock: And so it’s a reflection of who they are as a brand and as a company and as people.
David Brancaccio: Yeah, this looks sort of like a periodic table. Only one of the elements is HF for Half and the W for Wit. I get it.
Dawn Hancock: Yeah.
David Brancaccio: So, no machine-style mass production of designs here. Hancock tells me her effectiveness as a graphic designer is about listening carefully to her mainly socially-conscious clients and letting ideas ferment inside this hive of creatives working in Chicago’s Humboldt Park neighborhood.
Dawn Hancock: I don’t fear robots taking over our jobs anytime soon.
David Brancaccio: That’s borne out by a recent study of automation by the McKinsey Global Institute, the one we’ve been using all along in this series. It finds that just 11 percent of the work of a graphic designer can be automated with present technology. No fear of of a robot invasion immediately, but look to the future, and the crew at Fireberry gets jittery. The guy programming Firebelly’s website thinks the coding part will be taken over by machines eventually. He’s probably right. And listen to Tom Tian, one of Firebelly’s designers. He’s thinking meta, capital “M” about what happens if the march of technology someday reduces the amount of people needed to keep the economy going.
Tom Tian: I mean after all, robots don’t get married, they don’t need wedding invitations, they don’t start new businesses. Those things do account for a significant amount of economic activity that under-gird designers lifestyles, so if that kind of market doesn’t exist, then obviously it does affect us.
David Brancaccio: Tian may be onto something important that futurists and some economists have noted. Machines aren’t really consumers, in the case of robots, they mainly consume, what, electricity? They don’t buy new clothes every spring or go on interesting vacations. So what happens to the consumption-driven U.S. economy, not tomorrow but perhaps in the lifetime of a millennial, if many workers are machines with no outside interests or needs? That’s a key question to think about now, before the algorithms and machines are everywhere. This came up in the first podcast, but I’ll say it again: People currently get paid $2.7 trillion in wages for doing the type of work that lends itself to automation. That type of work accounts for nearly half of all work activities that humans do now. What happens when the tech advances and robots are in every office cubicle, behind every steering wheel, on every factory floor?
I’m David Brancaccio. This is Robot-Proof Jobs, the special podcast series from Marketplace, chapter three, the final bit. It’s an audio journey to consider how we humans might position ourselves on the job given all the technology that’s getting hired. The White House keeps talking about pushing back against globalization, but all these robots, algorithms, and artificial intelligence driven machines may be the next big threat or the next big opportunity. I keep reminding myself of the stakes here. Remember when Ro Khanna, the Silicon Valley congressman said this?
Representative Ro Khanna: We can’t as a country find common ground if folks are left behind from the technology revolution.
David Brancaccio: He’s the politician who wants to be sure that the Rust Belt, Appalachia and beyond gets in on the tech boom. It’s the reason why I went on this robots journey in the first place.
Representative Ro Khanna: If we want to have any chance of healing this country, bringing people together, we have to provide people with a concrete pathway to be productive in this new economy.
David Brancaccio: This new economy is where I’m taking this third episode. So far in this series, we’ve met a string of people smart enough or lucky enough to hold robot resistant jobs now. Dawn Hancock, the graphic designer in Chicago is just one example. Now, I want to look further into the future to see what happens when technological changes could go from incremental to revolutionary.
When I started poking around, I learned that some of the top minds in America are worried about a future where hardware and software can do a lot of the work humans do now. Their concern is driving them toward some fascinating, if not radical solutions, to help robot-proof the system. It’s a system that’s apportioned money in society, since well, at least since the industrial revolution. Some experts say we have to consider the possibility that the need for people to work in America tapers down. So how then do we make a living?
Martin Ford: I think that we can adapt to it in a way that people still have meaning even if perhaps they are not working as much in the future. But it’s definitely going to be a huge challenge both for individuals and for society I think to adapt to this change coming to us.
David Brancaccio: Martin Ford is a futurist whose most recent book, Rise of the Robots: Technology and the Threat of a Jobless Future, won the Financial Times-McKinsey Book of the Year award. We heard from him just briefly toward the beginning of this podcast series, about what he called “optimism bias.” Remember that?
Martin Ford: We always assume absolutely that technology is going to be a good thing for us and historically it has.
David Brancaccio: Keyword “assume.” The investment research company Forrester did a study predicting automation would create nearly 15 million jobs over the next ten years. So far so good. But the same study predicts that automation will wipe out nearly 24.7 million jobs. Do the math and that means nearly ten million jobs vanish. That’s like what happened during the great recession, when unemployment soared above ten percent. Here’s J.P. Gownder, the lead researcher on the study.
J.P. Gownder: It’s one of those boiling frog scenarios, where you put the frog in a pot of water, you turn up the heat slowly, and the frog doesn’t realize until it’s too late. And you know, that is part of this equation. We have to have public policies and such that will help to make sure other kinds of jobs fill in that gap of nearly ten million jobs.
David Brancaccio: And that’s by the time a college graduate this year turns 32. Beyond ten years, the effects could compound. Here’s Ford again.
Martin Ford: If you are facing the same challenges or if your job is even a little bit boring from day to day, then you should be a little bit worried that at some point technology is going to, you know, reach the point where that’s going to be threatened. But in the longer run, getting out beyond ten years or so, or 15 years, it’s hard to really say anything is going to be totally safe because really we’re seeing some pretty remarkable advances especially in the area of artificial intelligence called deep learning, where they are actually building systems that kind of loosely and at a primitive level really begin to model the brain and really are beginning to exhibit the initial stages of actually thinking. I think it’s going to have a huge impact.
David Brancaccio: Ford uses the analogy that machines in the workplace are like aliens who land on earth and are happy with work, work and more work. I asked him about that.
Martin Ford: I mean the only thing they need is food and water. They don’t, you know, they’re not going to go and buy an iPhone or an iPad.
David Brancaccio: See, that’s the economic key to that right, because if they don’t consume, what’s going to drive the economy?
Martin Ford: What I’ve suggested is that we ought to adapt our system to reflect this new reality and basically that means replacing jobs as the income distribution mechanism with something else. And I think the most obvious thing to do is to have some kind of a Basic Income so that everyone will get an income whether they work or not. This is something that many people find very offensive, but I think it’s probably inevitable that we’re going to have to move in that direction.
Chapter 1: Universal Basic Income
Robot-generated voice: Robot-proof option number one, Universal Basic Income.
David Brancaccio: Right now, some Silicon Valley technology people have funded an experiment to see what happens when you simply give money to people.
Matt Krisiloff: It’s done in a way where people could work, not work, but everyone receives an equal base playing field.
David Brancaccio: $1,000 a month in this initial round of research. The idea is to study how Universal Basic Income gets used in a real situation in Oakland, California. Matt Krisiloff is director of Y Combinator Research, part of the famous California tech incubator that does seed money for startups in innovative ways.
Matt Krisiloff: Over the course of 20, 30 years, we could have software that basically can think like a human can, we think it’s quite possible that a lot of jobs will become automated. And so we are trying to think, if that type of scenario happens, what can we really do to expand the social safety net and really meet people’s needs as our world changes.
David Brancaccio: It’s an idea that attracted some interest from both the political left and the political right. On the conservative side, economist Milton Friedman found merits in this, calling it a “negative income tax.” While voters in Switzerland not long ago voted this down, they have a kind of Universal Basic Income in the Basque region of Spain. And let’s not forget that in Alaska, people get checks drawn from profits on all that oil, just for being an Alaskan. Y Combinator researchers want to know among other things, whether free money causes people to seek out more fulfilling work as opposed to doing a job that’s soul crushing. Also, whether the money gets frittered away or robs ambition.
Matt Krisiloff: It’s such a big shift in how the safety net would be constructed and really for any country’s context, it’s really a tremendously expensive idea. Se we need to make sure it’s actually a good one first before actually advocating for something like this.
David Brancaccio: Some of the biggest of technology tycoons who hope their own innovations are changing the world for the better also seem deeply worried about what tech could do to jobs in the long haul. Tesla founder Elon Musk talks about Universal Basic Income. Yet, you heard the Y Combinator guy calling it “tremendously expensive.”
Matt Krisiloff: You know, if we jump to the endpoint, which again is pretty far out, if the real need for this is something like, we have tremendous automation, it’s hard to imagine just how much more wealth we could be creating as a society, it could be double, could be 1,000 times the GDP we have now if we really have infinitely replicable software that can basically just do anything that a human can at very, very little cost. But this type of technological change and this type of labor force change happens in a very gradual process. It’s easy to just imagine towards the end of, oh the robots have taken the jobs or something like that. But there are a lot of these forces that are very relevant right now, a lot of people out there right now that don’t have all the resources they need.
David Brancaccio: To do this, some economists suggest building on a mechanism that’s already in place to get money into people’s hands. That is, dramatically expand the federal Earned Income Tax Credit now targeted toward poor working families. A conservative think tank sees this as a replacement for social programs, for welfare. Liberals see it as a supplement to those programs. The Y Combinator study of Universal Basic Income is in its pilot phase. Meantime, what else?
Chapter 2: The Robot Tax
Robot-generated voice: Robot-proof option number two, the robot tax.
David Brancaccio: Listen to who was toying with this idea in public recently. Bill Gates himself, talking to the publication Quartz about taxing automation to offset the social costs of tech.
Bill Gates: Right now if a human worker does, you know, $50,000 worth of work in a factory, that income is taxed. If a robot comes in to do the same thing, you’d think that we’d tax the robot at a similar level.
David Brancaccio: Quartz put that music in there, by the way. Mr. Gates, who needs no introduction, says we could then take that robot tax money and funnel it into education and retraining programs for displaced workers, that’s what experts call them by the way, people who are pushed out the job, displaced. Bill Gates’ people sometimes are willing to get the great man when I call, but not this time. You know who did take my call? Another high-powered person, but one doesn’t like this idea of a robot tax one bit.
Lawrence Summers: Why would one single out among possible technologies, robots?
David Brancaccio: Former U.S. Treasury Secretary Larry Summers, who’s also a Harvard economist.
Lawrence Summers: At the practical level, why robots? Word processing programs displace secretaries, dishwashers — the automated kind — displace dishwashers the human kind. Electricity means many fewer people in jobs carrying things from one place to another.
David Brancaccio: Summers also has a philosophical beef with this idea, one he called an ultimately more important problem with taxing robots to offset their social cost.
Lawrence Summers: Should it be the objective of policy to encourage more rapid technological improvement or should it be the objective of policy to retard technological improvement? And it’s always seemed to me that enlarging the pie as much as possible, and then figuring out a set of policies that are directed at making sure that that pie is allocated in a fair way, is a much better strategy than seeking for some other reason to slow the growth. I think the best course is not to be an ostrich and pretend we can ignore disruption. Too many people do that, I was shocked and appalled when the Treasury Secretary said that artificial intelligence destroying jobs wasn’t on the radar screen of the administration because it was 50 to 100 years off.
David Brancaccio: Steven Mnuchin said that the other day, it was quite striking.
Lawrence Summers: I thought that was an extraordinary and clueless statement, but having said that, I don’t think simply trying to resist or stop technology is a viable strategy. And by the way, I don’t think it would be a viable strategy if it was adoptable on a global basis, but it’s even less viable for a single country in international competition.
David Brancaccio: Meaning we could end up losing work to places like Germany and Japan, which are not shy about using machines in industry.
Chapter 3: Micropayments for Your Data
Robot-generated voice: Robot-proof option number three, paying for your private data.
David Brancaccio: Some have proposed getting tech companies to make what are called micropayments. These tiny payments, lots of them, would go to people to make up for what they are losing in income as technology advances. Not for free, but as an exchange. If a social media company wants to follow our shopping patterns using our private data, maybe we agree to that in return for a small fee every time, a few cents here, a few cents there. It could add up. Tech companies purchase your personal data, the way you might sell blood for extra cash, if that were legal. Jaron Lanier, a computer scientist and composer who popularized the term “virtual reality” has promoted this micropayments idea. If folks at Google, Facebook or Snapchat are listening to this, they probably just got nauseous.
Chapter 4: Own the Robots
Robot-generated voice: Robot-proof option number four, buy stock in robots.
Richard Freeman: One of the things that’s happened in the last 20 or 30 years, the share of income that goes to capital has increased, and the share of income that goes to labor has fallen. We have to do something to change that if we’re going to have monies that allow normal working people’s living standards to go up. And they could fight against the capital, but fighting the machines makes little sense, so the best thing is that they should become capitalists, in some sense, and own part of it.
David Brancaccio: Noted labor economist Richard Freeman at Harvard has come up with this idea: We should all own the robots. He’s thinking about incentives so that lots more people own stock, shares of companies that will get richer off all the new tech, all the robots in the workforce. If productivity gains are going to shareholders not workers, why not find ways, policy ways, to turn a lot more workers into shareholders.
Richard Freeman: You certainly would want to have shares not only in the company that you are working at, that’s always risky, the company can have problems, but you’d want to have a broad based set of shares.
David Brancaccio: There are certainly examples of employee buyouts of entire companies in which a lot of the employees end up owning a piece.
Richard Freeman: Yes, about ten percent of American workers are in employee stock ownership plan companies. And many firms including most of the high tech firms have either profit sharing and ownership schemes and stock option plans, those are done because people actually work harder and better and smarter when they have some ownership stake, but it also gives people this protection if their wages don’t go up because of technological or other reasons.
David Brancaccio: Freeman says companies could get tax breaks if they pushed stock to employees. Workers, for their part, could also get a tax break for robot dividends. What this program needs is a catchy name.
Richard Freeman: I mean we’ve tried giving names to this, inclusive capitalism, share economy, but maybe we are too academic in our analyses to make it a good sales point. I’m not sure if the salesman that’s in the White House, if he’s listening, maybe he could do something good for everyone.
Chapter 5: Education
David Brancaccio: A Universal Basic Income, a tax on robots, selling your privacy for money, more universal stock ownership, and in our search for next generation policy approaches, let’s not forget this pachyderm that’s been standing here all along.
James Manyika: So typically when this conversation comes up in policy circles the the conversation usually runs to one place education.
Anca Dragan: That requires first education.
Erik Brynjolfsson: It’s not a matter of just investing more in education. All of that wouldn’t hurt, but it’s really more fundamentally reinventing education.
Michelle D’Attilio: Obviously to me, that change has to be an education.
Car radio: The one-0, misses up and in, two balls no strikes. Brewers up five to two.
David Brancaccio: The apex of my robot-proof jobs road trip was Fond du Lac, Wisconsin with over a thousand miles on the trip computer since my start Pennsylvania. I’d come here to talk to Ali Shrih at a civil engineering firm called Xcel.
Ali Shrih: I am excellent at math, James?
David Brancaccio: He’s just checked with his boss, also sitting with us to make sure he agrees and, it’s confirmed, everybody here is good at math. The McKinsey data we’re using says civil engineer is a very robot-proof career with just 13 percent of that job automatable, robotizable. The economy needs lots of them, there are more than a quarter million civil engineers in the U.S. economy. And it takes a lot education, including mastering the math, to become one.
Ali Shrih: Replacing an engineer would be much much harder than replacing a worker in a factory. And we have a lot of variety in our work that guarantees me a job I think, I hope, in the short term, until we retire.
David Brancaccio: Shrih designs the mysterious inner structures of concrete slabs for commercial buildings. These are precast in remote locations and trucked to the construction site. He tells me about software under development that could eventually let him put parameters into a screen here in Wisconsin that will be sent straight to machines at far-off factories. What this does is cut out the human workers at the factory, the fabricators. We say his job is safe for now, but Shrih has an imagination for disaster that suggests to him that given the speed of innovation, no job is completely robot proof. This is what he advises the engineering students he’s taught at the University of Wisconsin.
Ali Shrih: We have our minds always. We can always innovate, make things more efficient, smaller, faster, stronger. We have that on our side, yes.
David Brancaccio: Some places are better than others in trying to think ahead about developing the minds, the talent the skills that will be needed. In Australia, the state of New South Wales did its own local version of a robot-proof jobs study, so the training keeps up with the what companies want to hire.
In the U.S., the National League of Cities has just issued a call to action, commissioning a report on the future of work in cities given societal shifts and, yes, advances in technology. Among the conclusions, quote: “While the benefits of the new economy may largely accrue in cities, they will also likely be the epicenters of potential tensions and upheavals associated with automation.” Bill Peduto, the mayor of one of those cities, Pittsburgh, calls his town these days “Roboburgh.”
Mayor Bill Peduto: We can’t wait and see what happens, we have to work proactively as it’s happening to make sure that there are opportunities and benefits that are there for everyone.
Car radio: Accurate, dependable and late breaking. This is local and state news on WLEN, Lenawee’s local news station.
David Brancaccio: On the road trip, I had come through Lenawee County in Southeastern Michigan. The School District has a tech center here that pulls together the vocational kids and the college-bound math and science kids from all the high schools into one campus for one-half of every school day. Here’s where I found some students setting themselves up to be, I gotta to say, pretty robot-proof.
Caleb Christensen: I love talking. I’m definitely not afraid of a camera or a microphone that is.
David Brancaccio: There’s Caleb, who might be a future Ryan Seacrest, producer but also on-camera.
Caleb Christensen: I’m planning on going into the entertainment business.
David Brancaccio: Followed by Emma, who’s interested in something with just a 21 percent automation potential.
Emma Freshcorn: I want to be a pediatric nurse practitioner.
David Brancaccio: And this guy, Mitchell, who wants to cross swords with evil hackers.
Mitchell Sielsky: I’m going to be going into cyber security and network penetration testing.
David Brancaccio: I’m not even going to look that up in the robot-proof jobs database. If Mitchell gets good at thwarting cyber criminals, I’ll personally guarantee he’ll have as much work as he ever wants to do, in perpetuity. The stars like these always stand out, yet some guidance counselors who agreed to meet me at the tech center here want us to be realistic about how far they can push kids toward careers favored by the statistics.
Dana Noel: I think the greater concern for me is that I want someone that’s interested, a job that they’re interested in, because I would like them to be involved in something that they’re going to last at and not be quick to be done with it. So I do a lot of talking about their interests and where their skill levels are.
David Brancaccio: Kids are kids, and sometimes you just hope they grab onto something, anything, to motivate their learning. And for some, thinking too deeply into the future may be a luxury they can’t afford right now.
Karen Kelly: I met with a parent yesterday, who she wanted me to make sure all her child’s teachers knew, they have shut off notices for everyone of their utilities. The dad’s lost his job, you know, there’s hard times still.
David Brancaccio: Among the many reminders on this Robot-Proof Jobs quest that when it comes to education, training, re-training, easier said than done. Some places are better than others at it, but where’s the revolutionary approach that will allow training to keep up with the revolutionary changes we’re seeing with tech entering the workforce? I was looking for something big.
Chapter 6: A Manhattan Project
David Brancaccio: For me, it was a single phone call that changed my outlook on this completely.
David Brancaccio: Mr. Kalil, David Brancaccio in New York, thanks for taking the call at such time that you should be relaxing.
David Brancaccio: I was bugging the man while he was on vacation. Thomas Kalil was until this year Deputy Director for Technology and Innovation of the White House Office of Science and Technology Policy.
Thomas Kalil: Oh, happy to do it. It’s an important topic.
David Brancaccio: Kalil is now Entrepreneur-in-Residence at the University of California Berkeley. He had a question for me.
Thomas Kalil: The R&D budget of the Department of Defense is around $73 billion. Do you have any guesses as to what the Department of Labor’s research budget is?
David Brancaccio: Department of Labor, research. I don’t know, a little smaller?
Thomas Kalil: $4 million.
David Brancaccio: 4 million with an “m”, not a “b.”
Thomas Kalil: Yes yes. And so there’s not a group at the Department of Labor who’s saying how could we take advances in artificial intelligence and figure out how that would be used to reduce the time for a non-college educated worker to gain a skill that is a ticket to the middle class.
David Brancaccio: The point here is why isn’t the government investing more massively in new ways to help people jump onto this high tech jobs bandwagon? So what’s an example, then? Where, how and when has this type of thinking paid off before? Here’s the story: About ten years ago, the U.S. Navy was desperate to upgrade its training.
Thomas Kalil: The problem that they were trying to solve is could they reduce the time required for new Navy recruits to become an expert in a technical subject, and reduce that time from years to months.
David Brancaccio: They needed a lot of people to learn how to do Information Technology systems administration.
Thomas Kalil: This is a very important problem for the Navy because as you can imagine, if you were in a ship or in a submarine and you have a blue screen of death on your IT systems, that’s really a problem, and so you need people who are on the ships, and who are on the submarine, that can diagnose and fix your systems.
David Brancaccio: So the Navy got the Defense Advanced Research Projects Agency to look into this. Those are the folks who invented the internet. DARPA researchers studied the best way to train people quickly in a technical skill. They looked at the relationship between mentor and student.
Thomas Kalil: People have thought for a long time that a one-on-one tutor situation would be great. You know, if everyone had their own Socrates or their own Aristotle.
David Brancaccio: One-on-one, person-to-person, works, but let’s face it, it doesn’t scale very well. So DARPA starts modeling the best tutor-to-novice interactions, and a private contractor in Silicon Valley, a firm called Acuitus, comes up with a screen based artificial intelligence system to supplement human instructors. It’s got a pretty military sounding name, “Education Dominance.”
Thomas Kalil: First of all, they can figure out how to keep you on the knife edge between you being bored because the problem is too easy and you being frustrated because the material is too hard — so that’s number one. Number two, when you do get stuck, a good tutor doesn’t just give you the answer, they give you the minimum hints necessary to figure it out for yourself.
And then they started evaluating it, and what they discovered was that after four or five months, these new Navy recruits were outperforming people who had been with the Navy for seven to ten years. That’s when I got pretty excited about it, when I started seeing the results.
David Brancaccio: In other words…
Robot-generated voice: Robot-proof option number five, a massive astronauts-on-the-moon style research and development effort to push people up the ladder of the high tech workplace.
Thomas Kalil: For example 30 million Americans are reading at a third grade level or below, but you don’t see a huge effort on the part of the private sector to solve that problem. We have not thought seriously about how we would harness science, technology and innovation to advance economic and social mobility and create more ladders of opportunity. In part because the private sector is under investing because they may not see an immediate opportunity, and the the agencies that are responsible for worrying about these issues like the Department of Labor or like HUD, we’ve never said, hey you should have a research arm that could do for economic and social mobility what DARPA’s does for the military or what NIH does for biomedical research.
David Brancaccio: This is a show-stopping notion. A Manhattan Project to prepare the human workforce for what the economy needs as more tech comes online. Approach the smartest minds, the cleverest entrepreneurs and say here’s money to help you fix this. Here, solve this mismatch between what people know how to do and the work companies need to get done. There’s no need to start from scratch. Among others, the Advanced Robot Manufacturing Institute we heard about in episode two is onto this idea, in part inspired by Department of Defense money.
So is now the time to pull the trigger and get going on some of this? Maybe. Maybe not. Marketplace and Edison Research did a public opinion poll this spring. Just 28 percent were worried about automation and robots taking their job. Plus, it’s not what the politicians are talking about.
Steve Mnuchin: I think that is so far in the future, in terms of artificial intelligence taking over American jobs, I think we’re so far away from that that it’s not even on my radar screen.
Here is that moment that Larry Summers at Harvard was shaking his head about a few minutes ago. In DC this spring, the current U.S. Secretary of the Treasury, Steven Mnuchin, was asked a question about artificial intelligence taking jobs. He did not seem worried.
Axios’ Mike Allen: How far away?
Steve Mnuchin: Far enough that it’s…
Mike Allen: Seven more years?
Steve Mnuchin: Seven more years? I think it’s 50 or 100 more years.
David Brancaccio: Look, maybe it’ll be 50 or 100 years before an artificial intelligence is appointed U.S. Treasury Secretary. And maybe in prepping for his new job, Secretary Mnuchin might not have seen the report issued by the outgoing Obama Administration that nearly half the jobs in America could be affected by AI or automation. And the Treasury Secretary, like all of us who ever took an economics class, would have studied that technology has tended to raise standards of living over the centuries. But as I keeping hearing from the experts, this time it may be different with the technology now developing so insanely fast. Machines can read X-rays and identify cancer cells. How long until I get to announce on the radio some morning that a robot car just drove itself coast to coast with nary a bathroom break?
Chapter 7: Robots are Everywhere
David Brancaccio: You just never know where technology will show up on the job now. Here’s our last example of where robot technology is going. It’s called Benjamin, software on a laptop. Co-creator Ross Goodwin and I get typing.
Ross Goodwin: So maybe give me a sentence about art.
David Brancaccio: Art is everywhere.
Ross Goodwin: Maybe a little longer than that.
David Brancaccio: Art is everywhere in our natural environment.
Ross Goodwin: Perfect, art is…and then I’ll just let the machine complete.
David Brancaccio: “A difference in manner and in a profound production of art and beauty and in the sense of the artist life…
David Brancaccio: Benjamin, the software, learns from the writing of humans to create original stories of its own. If you’d like to construct a writer robot like this, it’s child’s play. Here’s all you do.
Ross Goodwin: It’s called a long short term memory or LSTM recurrent neural network, and its an algorithm that’s been around for about 20 years or so. But it only recently started to work well because of advances in hardware, particularly what are called graphics processing units, GPUs, which are used for gaming but also for data science, large scale data science.
David Brancaccio: At the time we met, Goodwin was about to embark his own road trip from New York to New Orleans with ol’ Benjamin here to see if visual inputs might inspire new stories to spill forth. And know this about Benjamin, this software was fed the scripts of hundreds of movies, and by itself generated a movie outline.
Oscar Sharpe: Subterraneum artist, a comedy about a young man who is at the center of a competition for a young girl, who wants to prove he is a man of sexual activity. He is also a professional star and a part of the camera…
David Brancaccio: Reading the outline there is Oscar Sharpe, a human filmmaker and Hollywood screenwriter who works with Goodwin. Real actors later interpreted this synopsis and the resulting short film created quite a stir at a science fiction film contest.
Oscar Sharpe: Not only was it short-listed in the top ten with some quotes from the judges, the human judges, that were pretty delightful, like, “the only thing that would enjoy this is another computer” and “I’ll give it top marks if they promise never to do it again.”
David Brancaccio: Here’s a clip from the result.
Movie clip: What are you doing…I don’t want to be honest with you…You don’t have to be a doctor…I’m not sure, I don’t know what you’re talking about…
David Brancaccio: As sketched out by a machine. Goodwin and Sharpe are emphatic that this is not a replacement for a screenwriter, an instrument that humans can use to play a new riff, like a saxophone. But you can see where this is going, this technology goes eventually goes everywhere. Just a decade ago there were no iPhones on the market. Imagine the technology in another ten years. The question is, will we be ready? What do you think?
How robot proof is your job? Marketplace.org/robotproof to look it up. Also photos of the road trip, the robots, the people we met. The producer of this Robots-Proof Jobs podcast series is Katie Long. The Senior Producer is Nicole Childers. Clare Toeniskoetter, Paulina Velasco, Ben Tolliday, and Daniel Ramirez worked hard on the mix. I’m David Brancaccio. This is APM.
Note: Transcript may contain typos or errors. Please cross-reference with the corresponding audio to ensure accuracy.
There’s a lot happening in the world. Through it all, Marketplace is here for you.
You rely on Marketplace to break down the world’s events and tell you how it affects you in a fact-based, approachable way. We rely on your financial support to keep making that possible.
Your donation today powers the independent journalism that you rely on. For just $5/month, you can help sustain Marketplace so we can keep reporting on the things that matter to you.
