Bill Kerr: Welcome to the Managing the Future of Work podcast from Harvard Business School. I’m Bill Kerr. The following is a guest appearance—one of an occasional series of podcast episodes produced elsewhere that highlight research and expertise from the Managing the Future of Work Project. My project co-chair, Joe Fuller, recently appeared on the podcast the Q Factor to parse future-of-work trends for an audience of investors and business leaders. In this wide-ranging conversation, Joe sheds light on AI’s impact on work and hiring, the emerging gig marketplace for high-skilled professionals, remote and flexible work, the importance of social skills, and much more. We hope you enjoy the discussion.

Gregg Fisher: The nature of work is changing. Rapid technological change, shifting global product and labor markets, evolving regulatory regimes, outsourcing, and the fast emergence of the gig economy—all of these factors are coming together at an unprecedented pace and scale. How can leaders prepare their organizations and create the workforce of the future? That’s today on the Q Factor.

Welcome back to the Q Factor. I’m Gregg Fisher. What does the future of work really look like? There’s no one better to address this question of the future of work than Joseph B. Fuller, Professor of Management Practice in General Management at Harvard Business School and co-lead of the school’s initiative Managing the Future of Work. Joe was a founder, first employee, and longtime CEO of global consulting firm Monitor Group, now Monitor Deloitte. His research interests encompass multiple issues related to the future of work, including the skills gap, the growth of gig work, and the care economy. I’m thrilled to welcome Joe to the Q Factor today. Here’s our conversation. Professor Joseph Fuller, welcome to the Q Factor.

Joe Fuller: Gregg, pleasure to be with you.

Fisher: It’s so good to be with you, Joe. So your signature project is Managing the Future of Work. It sounds like a project that could have been born in the pandemic, when we all left the office. But your work on this subject began much earlier, around the 2008 financial crisis, which is when we first met. I remember being at Harvard, attending a lecture that you were hosting, and hearing you talk about the future of work. And I’m sitting in the back of the room thinking to myself, “What the heck is the future of work? What is this guy talking about?” And, gosh, who would’ve imagined, you know, here we are 10, 15 years later, and we’re in the future. Like, we’re living it. We’re doing what you were talking about. You were really early to this set of ideas. So tell us the story of the genesis of this project for you. You know, during those first few years, what were your most important initial findings and learnings? In those early years, was your goal purely research? Was it knowledge based? Was there a more practical end goal in mind? Were you hoping to create some future for all of us, a path forward? Tell us, give us the story.

Fuller: Well, it’s not very dramatic, but as you may remember, Harvard Business School had a large project about the competitiveness of the U.S. economy that had started in 2009. And when I joined the faculty a couple of years later, I noticed that there was a variable that had been isolated through some survey work we’d done with our almost 100,000 global alumni that indicated that their opinion was that the workforce in the United States had been a significant source of competitive advantage historically but that that advantage was waning rapidly. And in joining the faculty that were leading that competitiveness project, I kind of raised my hand and said, “What do we make of this? And what do people know about it?” And frankly, what came to light is that there wasn’t really anybody at Harvard Business School who was researching that. So I raised my hand to try to understand what was going on with that. And what we came to understand is that, yeah, there were real deficiencies in the skills development system, the education system, in the United States, but that a lot of the problems companies were observing and, frankly, complaining about were often substantially of their own making—that the way they approached identifying candidates, the way they screened candidates, the requirements they put on candidates effectively created an artificial shortage. And a leading example of that would be a paper we did about five years ago called Dismissed by Degrees, which just showed the impact of requiring a college degree of an applicant for a job.

Fisher: Right. Because I guess the idea would be that, you know, people wouldn’t even get interviewed. They wouldn’t, you know, even get to the table if they didn’t have a college degree in most cases.

Fuller: Exactly. And what we found is consistently, Gregg, that companies tried to make their recruiting process as efficient as possible. So a good-size company offering a job that paid more than a $60,000 salary a year before Covid would get 200, 250 applicants, on average, and would interview three or four people total. And shrinking that pool from 250 to three, you’ve done almost exclusively by a technology called an “applicant tracking system,” which is, it is based on AI, although I don’t really think it’s all that intelligent. That’s maybe giving AI bad name. And it uses very blunt proxies to evaluate the attractiveness of workers.

Fisher: Like a college degree: Yes/No.

Fuller: Yes. Or a criminal conviction: Yes/No. Or some more subtle ones. One that’s particularly insidious is called the “continuity of employment filter.” For about 50 percent of middle-skills jobs in the United States—a middle-skills job is a job that requires something more than a high school diploma, but less than a college degree, so some type of certificate, license, study of some topic that would be specialized—50 percent of employers drop a candidate completely from consideration if they have a gap on their resume of more than six months. And you can see, you know, would why an employer do that? You could imagine they’d be saying, “Well, maybe this person doesn’t have a lot of get up and go, a lot of self-efficacy. It doesn’t sound like former bosses or colleagues were standing in line to try to hire this person when they work their network.” So you could say, well, maybe it’s a less attractive candidate. And something to always keep in mind about hiring is, it’s a relative phenomenon. It’s not, you know, is Joe qualified? Or is Joe a superior candidate or an inferior candidate to Gregg? If you think about six months, yeah, maybe that’s a long period, but if you’re having a problem pregnancy, you might well be out of the workforce for six months. If one of your parents were terminally ill and you had to go care for them, if you’re a military spouse and you got transferred from 29 Palms to Okinawa, six months is going to go by in the bat of an eye. And, of course, once you’re outside that strike zone, it starts to compound, because you’re now not getting considered for half the jobs that you might be qualified for, your odds of getting the next position fall, and you can get a compounding effect. So we regularly found that employers didn’t really understand the systems effects of the way they went about it.

Fisher: Interesting. I find it amazing with all the technology we have today that, you know, this whole process still is pretty antiquated. I think about reviewing people’s resumes. I was having a conversation with a young person looking for a new opportunity recently, and he said to me, “You know, I’m concerned. I’ve only been in …” he had, I think, three jobs, and he was about 30. And he said, “I’m concerned. I’m in this current situation now, and I’m really unhappy, but I’m afraid to leave, because I don’t want my resume to look bad that I’m jumping around too much.” And I thought that seems so inefficient. There’s so much more data that we have on this person. There’s so much more that we could know before meeting him. Why would he have to make a suboptimal choice about his future, being miserable in a job because you don’t want your resume to look bad, it just seems so weird to me.

Fuller: I’m afraid that there’s the good news and the bad news there. The bad news is that this person really didn’t like their job and felt compelled to stay in it for a while. The good news for him is he was smart enough to understand that the AI isn’t very smart. And, therefore, he was right to be concerned. And just to give you another example of a hack, there are firms that will charge you for advice about how to game these applicant tracking systems in order to maximize your chance of getting an interview. And one thing they do is a trick that’s kind of wonderful in its deviousness is, they tell people, copy the job description, put it in white ink so it won’t show up on a screen, and paste it into your experience.

Fisher: Oh gosh.

Fuller: So the AI can see the white ink, and it says, “My God, this person is perfect.” The human being that gets the printout or looks at the screen can’t see the white ink, [can’t] see that you’ve hacked it, and has to rely on their own wits and their own intelligence to detect that maybe you’re not all that the AI claims you are.

Fisher: There are these AI researchers, Katja Grace and John Salvatier, they did some research and a report, and they had claimed that there was a 50 percent probability that humans would be technologically redundant by the early 2060s. So they were sort of taking a different angle on this to say that, you know, the AI and the technology was going to put more of us, I think, out of work or make us redundant. But that other argument is, instead, it creates these new opportunities. I think one of the things I was reading about, you know, a lot of times when we think about robots and technology taking our jobs away, we think of these manual routine tasks.

Fuller: Yes. Right.

Fisher: But now, you know, you hear about how AI is doing a better job with medical diagnoses or giving legal advice. They can very quickly go back and read all the literature in seconds of every time this thing has been looked at before. And in my mind, that actually sounds pretty good. I think I’d prefer that over one particular human being and their own personal experience. What are your thoughts on all that—this redundancy concept?

Fuller: There’s a difference between theoretical redundancy and redundancy that actually expresses itself in a marketplace.

Fisher: Right.

Fuller: So could you automate rice farming in Southeast Asia by having highly nimble robotics doing the planting, going through those rice patties with AI predictive algorithms, talking about blight or insect infestations and recommending fertilizer? Sure you could, but all adoption of technology is governed by fundamental substitution economics. And the economics of substitution are, what does it cost to substitute—in this case technology for labor—and do I get the same value, or do I get a higher value or a lesser value? Where AI is making unbelievable progress is in what would be defined as non-routine, cognitive jobs. So it’s cognitive in that it’s intuitively obvious what we’re talking about; but it’s non-routine, which means there’s a fair amount of variability around the task. What’s a classic routine job? Assembly line worker; just do the same thing exactly the same way every time, and don’t add value. Right? But a lawyer or a radiologist has historically done a non-routine cognitive job, because each X-ray or MRI scan is different. Now, can we imagine a world where the human is completely out of the loop in radiology? Yeah, we could imagine it, but I don’t think it’s going to happen in the foreseeable future. I think that radiologist is sitting there pretty soon—way sooner than 2060—with imaging that’s been digitally enhanced using AI with an AI diagnosis, and they’re checking it more than they are arguing with it. But, you know, AI is not independently intelligent. It relies on structured or unstructured learning to understand things. And it’s not good at recognizing the one in 1,000, the one in 500, exception. It may not also be very good at, for example, saying, it’s recommending some, you know, very significant surgical intervention, because it doesn’t realize the patient is 98 years old, it doesn’t realize the patient has mental health issues, but that has to factor in. But what is coming is for supply-chain managers, for routine contracting in law firms, particularly what’s called “cognitive AI,” companies like Aera Technology. I mean, what they can do now is really rather remarkable. And that means you’re going to get technological displacement in good-paying white-collar jobs that have been more or less impervious to historical disruptions caused by things like globalization. And that kind of displacement may add more wood to the fire, pour some gasoline on the fire, for some of the associated political dislocations we’ve seen as people have come to the conclusion that the way the world has unfolded economically has been very much against their interests.

Fisher: When you talk to the various companies that you’re communicating with and thinking about this automation and AI and all that’s happening in the world and the future of work—as you know, I invest for a living. I’m fascinated particularly with small, fast-growing, new, innovative businesses—but are you seeing data that shows that all of this is helping companies to spend less money on labor, that their businesses scale more? Think about these gig platforms—like Fiverr, you mentioned earlier, or Catalent, which I know we’ve talked about in the past or others. Is all of this evolving—you know, AI, gig economy, these platforms, I can go hire a drone editor in two seconds for not a lot of cost. Is this reducing labor costs inside businesses that has them scale more and make their gross margins higher?

Fuller: I believe, yes. Has that been definitively proven? I wouldn’t say definitively. But that’s another effect, Gregg, which I think in some ways more important, which is, it creates access to world-class talent for employers that are going to need those skills but can’t afford the talent.

Fisher: Right.

Fuller: There’s a white-hot market for machine learning people right now, and if you are a third-tier industrial component supplier in the Rust Belt of the United States, and you are getting pressure from Caterpillar or General Motors or Boeing to be able to use machine learning as part of their broader management of their supply chain technologically, and you post a job in your hometown of Kokomo, Indiana, looking for a machine learning specialist, what are the odds that someone’s going to show up who is really world-class and want to work for your $100 million third-tier component supplier and live in Kokomo? Nothing against Kokomo. But maybe there’s a new resident physician at the local hospital whose spouse is a machine learning person. Maybe there’s someone who grew up in Kokomo or nearby that wants to, you know, loves the lifestyle there, wants to be near their friends from high school or the folks or whatever else. So it happens. But one of the strange things about the way technology is evolving is, historically people sought talent with a lot of experience in the vertical of their industry that the company worked in. Now people are looking for digital talent. So ExxonMobil is trying to hire the same person in terms of machine learning as is Capital One, as is Apple, as is the NSA, as is Merck.

Fisher: Right. And why would that person who has that very important expertise go work full time for one company without a whole lot of dough? They’re better off just being on a gig platform, and I could hire them and so could IBM.

Fuller: Exactly. And so I think, particularly for those middle-tier companies, for companies that could be, you know, terrific companies. Take a company like Whirlpool, Benton Harbor, Michigan, the last U.S.-owned white goods manufacturer—refrigerators, washers, dryers, whatnot. The western shore, the Lake Shore, of Michigan is beautiful. It’s a nice place to live. But Whirlpool is kind of the only game in town. And if you are coming out of even Notre Dame—which is probably an hour’s drive or an hour and a half’s drive to Benton Harbor, so pretty close—if you’re an engineering graduate coming out of Notre Dame or an accounting graduate, you would you rather go to L.A., New York, Chicago, Dallas than bet it all on Whirlpool? And I think Whirlpool is an exemplary company, an exemplary employer. I’m not picking on them. So the ability to source talent is hard to get, the ability to get world-class talent. It allows you to variable-ize your workforce costs.

Fisher: In a conversation about the future of work, we can’t ignore this idea of, you know, in-person versus remote work. There hasn’t been a conversation that I’ve had with a group of people in the last year or two where we haven’t thought about this. Even I, myself, and my own company now in New York City, I went from having a large, full-floor office with 80 employees coming in every day to, you know, my new business, which has a smaller team and a smaller office, and lots of flexibility on the work environment, and it’s working extremely well. And I guess we’re all wondering, like, what the future will be. Another colleague of yours, Tsedal Neeley, wrote a book called The Remote Work Revolution, which is a great book. And it talks a little bit more about how to succeed as maybe a leader of a business that has to embrace these ideas. I wonder what your thoughts are on this. Google recently said—and they’re, you know, a moderately smart company—they’re saying their employees can expect to be in the office, like, three days a week. That’s what they’re shooting for, versus Elon Musk, who basically took the other side of that and said, you know, remote workers are just pretending to work, and you all have to come in now. Then there’s the history of this. Like, this is not necessarily new, although I think the pandemic clearly opened up a whole new scope of opportunities, both for workers and for employers, to tap the global talent pool and all that we’ve talked about before. But you’ve got companies like Unilever—this huge company that, you know, they’ve been doing remote work for years. But some companies that haven’t, it’s hard to pivot. And then you’ve got, you know, new start-up companies like AppFolio that were born into this environment, and they’d never had it any other way. What are some of the things you’re seeing? I mean, is it easy for a large company to pivot? Do start-ups just start this way? What are your thoughts on this work-from-home versus come-into-the-office narrative?

Fuller: Well, it’s a very white-collar narrative. And we have to start with saying less than 40 percent of man hours in the United States can be done remotely.

Fisher: Right.

Fuller: I think that Tsedal’s work is very insightful and does give executives and decision makers an integrated way to consider the questions they’re facing. What I’m observing in working with companies is that a lot of companies, a lot of large companies—and some of them run by well-known alumni of our school—have been very anxious to declare some “new normal;” that we’ve thought of that, and here are the new rules. And my standard observation to them—to the point that some of them are getting tired of talking to me—is, we don’t know what the new normal is. Why don’t you just try to get to the next normal?

Fisher: Right.

Fuller: And why don’t you think about talking to your employees through a lens of consideration—you know, here’s what we’re going to try, but this is not a permanent policy decision. We’re going to see how this works out, and we’re going to decide in nine months how this is working out, or that is working out. It’s super difficult if you’re running a big, multisite company to come up with uniform rules. If you look at the big professional services firms, several of whom are clients of mine, what they see is, their attendance in their smaller offices or in secondary markets is 70 percent, and their attendance in major metros is 20 percent. And the difference is commute time. So if you’re in Milwaukee, and you’ve got a 20-minute one-way commute pretty reliably, versus you live in Berkeley, and you’ve got to get to San Jose every day, and you’ve got a reliable 80-minute one-way commute time, that’s a huge consideration. Not just about wear and tear in your job satisfaction, but on your productivity. So I think companies have got to start thinking about processes, not job titles. What are the processes that actually require some more physical interaction, versus those processes that don’t. This isn’t an executive directors and above should have to be here three days a week, but directors and managers and supervisors have to be four days a week. I mean, that’s all Taylor-ite thinking.

Fisher: It’s like a two-class society, right?

Fuller: And so it’s process driven. And over time, Gregg, where I think you end up—you know, it’s always dangerous to make predictions, especially about the future—but I think that, over time, better companies are going to have more variable-ization of work terms and more customization for important employees. And I’m not talking necessarily high-paid employees. Work is as much a sociological event as it is an economic event. And we know from data that people that have been hired into companies during Covid, who have not been to headquarters, haven’t gotten to meet people personally in real time, haven’t built deep social connections or even interests in other things going on in the company—the softball team, the Habitat for Humanity weekend, the support group for someone with some outside concern needs—that they have higher rates of turnover, less levels of engagement, with those employers. And I do think no one has an answer yet.

Fisher: This flexibility of work—you know, the employer having to offer flexibility—I mean, I’m a growing, small company again, and I’ve been hiring people, and the first thing, almost the first thing, that comes up in conversations is, you know, can we work from home? I mean, in order to compete to attract talent in my world—and I recognize it’s mostly like executive-level positions—but it has become a requirement. It’s no longer an option. Like, yeah, of course, you know, you’ll come in two, three days a week. If you were talking to someone, saying you’ve got to come in every single day, you’re probably not competitive any longer. And this other thing about the gig economy—and you know, we all have a skill that we offer for a fee, whether you work for a company and you have one customer, or you are in a gig platform and you’ve got many customers—either way, everyone’s an entrepreneur. And it’s an interesting way to think about these things. But I think about it for my kids right now, and I wonder, a lot of our listeners are people working in businesses that are mentors and have mentees, or maybe they have young children, like I do. And I’m thinking about, you know, what advice should we give our kids for the future? The great historian Yuval Noah Harari had said that career volatility will be much more significant in the future—in the near future—and our kids can expect to change, not just their jobs, but their careers multiple times. What advice or impact steps you would take? Or what should we be doing for the young workers today to prepare for the 21st century?

Fuller: Well, I think it’s a—once again, that’s one of those fundamental questions that does require a lot of thought. I’d add something to Harari’s point, which is, young workers expect that.

Fisher: Huh.

Fuller: What they are saying is, “I don’t anticipate working any place for more than four or five years. I don’t want to be like my folks. I’m perfectly comfortable going from full-time work to gig work back to full-time work.” I would say that young people have been not wanting to be like their parents since Cain and Abel, so maybe some of this is overblown. But the infrastructure to support that—whether it’s gig platforms like a Catalent or an A.Team or a Fiverr—or the liquidity of a labor market, where LinkedIn is looking for you, and you can look on LinkedIn. You know, when I graduated from Harvard Business School, if I was interested in interviewing at a company, I went and got the mailing address at the career-services office and sent them a snail mail.

Fisher: And you turned out okay.

Fuller: We’ll argue that one on a different podcast. But what does that mean? Well, the first thing for workers means that one of the most important skills for someone to hone and try to improve on, if they don’t come naturally, is a whole category of skills called “social skills.” That does not mean, by the way, life of the party, best beer-pong player, you know, tells great jokes. It can mean those things. There’s kind of a co-linearity. But it’s someone’s ability to have what’s called a “theory of the mind,” where you can … an easy way to think about is to put yourself at the other person’s shoes. Do I have the capacity to imagine what it’s like to be this customer, if I’m a sales rep in a mobile phone store, who has just broken their brand-new iPhone and is coming in to find out if the damage is warranty covered or whether they can still buy the insurance they turned down when they bought the phone. And I, in this particular scenario, have to give them a lot of bad news, but I can be empathetic. I can say, “I know that you must feel terrible. But you’re not covered. You didn’t buy this. And, you know, we can give you a quote on fixing it. But I want to caution you; I think it’s going to end up being more expensive to fix it than just get a new phone.” You can say that in a sympathetic, empathetic way. Or you can say, “No, you’re not covered. Hey, you didn’t buy the insurance. What were you thinking?” Or it’s also a spontaneous written and oral communication. So the ability for me to describe to my store manager what happened with that customer, who then got upset or said we were terrible, and we’re never going to use them, and that’s why I gave them a $100 coupon, saying, “Well, I know you’re upset, but it’s too bad. Your phone is broken, so here’s $100 off coupon”—to explain why I basically spent $100 rather to have that person go to a different company and change their subscription. Or I can write an email to that effect, and the recipient knows what actually happened, as opposed to trying to figure out what I meant by that. So now, unfortunately, social skills get disproportionately cultivated in early life. And we’re just beginning to understand how to upgrade someone’s adult social skills. It cannot be done to an infinite extent, but it can be done. A second is, people do have to have a familiarity with some fundamental technologies. Even if they’re not going to get a university degree, which will be a majority of the population, if they’re going to work in a customer-service function, you know, in a job that pays decently but is not going to make anyone rich and famous, they have to understand the basics of data. They have to understand the basics of digital technology. You can’t be a numerophobic digital illiterate and be productive in the economy of 2030. It doesn’t mean you have to be a mathemagician. It doesn’t mean you have to be a software engineer, but it means you’ve got to have certain basic capability like the old, you know, reading, writing, and arithmetic-type logic. You’re going to have to be able to respond in real time. You’re going to have to be able to derive data from visualization. You’re going to have to be able to understand basics of statistics. You’re going to have to be able to be comfortable understanding how increasingly user-friendly intelligent devices work. You’re going to have to be able to fight that tendency that, if it doesn’t work as you intuit it, you immediately assume it’s got a problem where you can’t figure it out.

Fisher: Thank you for sharing that. So, Joe, I have this tradition on the Q Factor. They’re called the “Three Qs.” And given all we’ve talked about, you won’t find this very difficult. It’s basically three questions that I ask every guest, no matter what their background or expertise. And here they are. So, number one, beyond economics, where do you see big data having the biggest positive impact on the world over the next 10 years? It could be anything: science, manufacturing, investing, philanthropy, global healthcare, climate change, whatever you think. Over the next decade, where and how will big data help the world the most?

Fuller: Public health. You know, I think that we’re beginning to get enough data and enough devices out there that we can anticipate lots of health events. And I point you to the data from the Narayana Hospital chain in India …

Fisher: Yeah. I’ve seen that.

Fuller: … where, you know, a quarter of heart attacks in India happen to people under the age of 40. And the types of devices we have just on, you know, Fitbits and Apple iPhones and whatnot to be able do anticipatory diagnostics. I think we’ve had a massive wake-up call on pandemics, and you’ll get a lot of—not just government investment—but I think you’ll get a lot of private-sector innovation in that space, because the health delivery systems in the world are not fit for purpose as they are now. And when you have a more aging population, and if you don’t—I mean, the public health system in China is a disaster. And there’s no question about, there’s enough purchasing power out there for innovative solutions that get economically effective results. So I hope that isn’t more hope than logic expressing itself. But that would be my answer.

Fisher: Yeah. Hands down, that’s been the answer that I hear most often, and I agree. And I’ve been stumbling onto all kinds of interesting businesses that are doing creative things using technology and healthcare. So I do think that’s a really exciting area. A second question is, really, on the flip side. In what aspect of the world do you see as the biggest, the most to be threatened by big data over the next 10 years? You know, where does big data pose the greatest threat in your opinion?

Fuller: That’s an interesting question. I think that it’ll threaten companies that rely on pricing and feature discrimination between their customers to make money. And I’m not talking about redlining or some kind of discrimination in a legal sense. I’m talking about the high-margin customers having greater awareness that they’re paying a high price relative to competition, that they’re paying for features they don’t use, that they are, you know, that they’re entitled to services that they’re not using. And if you get into a lot of companies’ economics, what you’ll find is that 15, 20, 25 percent of heavy users or price-insensitive users are driving 80 percent of profitability. And if those profit pools start getting exposed, either through big data to disruptors or their competitors or by big data to the user, themselves, a lot of companies are going to suffer as a consequence. And that’s up and down for everything from consumer financial services companies to professional services firms. And you can see data about this going back to things, you know, way back in the early days, of things like the advisory board, which companies started getting data, comparative data, with peers on their healthcare expenditures. And it really changes a company’s approach to that. Imagine that there’ll be maybe a company you fund or found that allows me to get in real time accurate and in some way certified or endorsed judgment about how, whether or not JPMorgan Chase is giving me the type of deal they could, they should give me.

Fisher: Right. Or am I getting the worst price across their entire customer base, or something like that.

Fuller: Exactly. And, and, and big data allows for more and more accurate and sometimes hyper-accurate transparency.

Fisher: That’s actually super interesting and different than what I’ve heard before. Thanks for sharing that.

Fuller: Glad to get an original one on the second one. Sounds like I came up with a bog-standard answer for your first one.

Fisher: No, actually, both were great. And it’s interesting. Your second answer makes me think about like, you know, does the 80-20 rule change because of AI?

Fuller: Yeah. Yeah, that’s a good way to put it.

Fisher: Well, the third and last question is AI—artificial intelligence: Is it our friend or is it our foe?

Fuller: I think the technology is our friend, and that we have to be very vigilant that misuse of it doesn’t make it an unintended foe. The opportunity for productivity, whether it’s releasing working capital or reducing scrap rates or deploying the right talent to the right project at the right time by anything from a government to a company, the potential of AI to unleash huge, huge waves of productivity growth, which will be strictly necessary in economies with no-growth labor forces, where the productivity of workers is going to be the only variable that’s driving GDP growth. You know, the potential is, I know it’s a little breathless to say it’s unlimited, but it’s tremendous. But as we can see, you know, in various states that trend toward totalitarianism, it can be abused in all sorts of ways. It can be abused to inhibit individual liberty. It can be abused in ways that discriminate between people, companies, types of companies. So it’s going to be, you know, like any tool, a bottle of … Alfred Nobel started the Peace Prize because he invented this great substance for things like mining and construction called dynamite, and suddenly it became high explosives in world wars. And we have to do what we can to ensure that AI does not, you know, stays in construction and mining and doesn’t make that Nobel transition.

Fisher: Right. Well, Joe, thank you so much for joining us on the Q Factor. I loved our conversation.

Fuller: That sounds great. It’s been a pleasure to talk to you. I enjoyed it.

Kerr: We hope you enjoy the Managing the Future of Work podcast. If you haven’t already, please subscribe and rate the show wherever you get your podcasts. You can find out more about the Managing the Future of Work Project at our website hbs.edu/managingthefutureofwork. While you’re there, sign up for our newsletter.