Later-Career Advantage in an AI World with Ben Zweig

Careers don’t peak and fade—they evolve.

In this episode of the HRchat Podcast, host Pauline James is joined by Dr. Ben Zweig, economist, data scientist, and CEO of Revelio Labs, to unpack what AI, remote work, and risk-off hiring really mean for later-career professionals.

The data-driven headline may surprise you: while AI exposure is dampening demand for junior roles, experienced roles remain largely untouched. Ben explains why this shift reflects a deeper transition—from task execution to orchestration—where coordination, prioritization, and cross-functional judgment become the most valuable skills in the economy.

Together, Pauline and Ben explore the difference between procedural organizations that automate easily and adaptable environments where roles evolve continuously. They discuss practical strategies like job crafting, choosing leaders who encourage experimentation, and navigating the loyalty tax without drifting into stagnation. The conversation also covers remote work’s “suburban advantage,” lower job mobility in risk-off markets, and why experience is increasingly rewarded when employers prioritize near-term delivery.

Ben also previews his new book, Job Architecture: Building a Language for Workforce Intelligence, showing how better job taxonomies and LLMs can bring clarity, speed, and fairness to people decisions.

If you’re thinking about your next career chapter—or advising others through change—this episode offers a clear, data-backed roadmap.

Key topics include:

• Why AI is cooling entry-level hiring but sparing experienced roles
• Adaptable vs. procedural organizations
• Orchestration as a core human advantage
• Job crafting as a hedge against stagnation
• The loyalty tax and mid-career earnings trade-offs
• Remote work trends and later-career opportunity
• Using job architecture and LLMs to structure skills and roles

Transcript

Welcome And Guest Introduction

SPEAKER_02 0:02

Welcome to the HR Tech Show, one of the world's most downloaded and shared podcasts designed for HR pros, talented Xebs, tech enthusiasts, and business leaders. For hundreds more episodes and what's new in the world of work, subscribe to the show, follow us on social media, and visit hrgazette.com.

AI’s Uneven Impact On Careers

SPEAKER_03 0:25

Hello, I'm Pauline James, CEO of Anchor HR and Associate Editor of the HR Gazette. I'm excited to be here today with another episode of the next chapter. For too long, careers have been described as if they peak and then wind down. But life doesn't follow a script, and neither should our work. It's my pleasure to introduce Dr. Ben Zwag, the founder and CEO of Ravelio Labs and adjunct professor at NYU Stern School of Business. Ben is an economist and a data scientist whose work sits at the intersection of labor economics, technology, and analytics. Through Ravelio Labs, he leads a mission to help organizations and society more broadly better understand the dynamics of the workforce by transforming labor market data into powerful insights. In addition to his role as a founder, Ben is deeply passionate about education and mentorship. At NYU STER, he teaches and inspires the next generation of data enthusiasts and business leaders, empowering them with the analytical tools to solve real-world problems. Ben, you've spent your career studying how the labor market is changing. From your vantage point, what stands out right now about how career paths are evolving, particularly in the later stages?

Adaptable Versus Procedural Work

SPEAKER_01 1:52

Yeah, first of all, thank you for having me and for such a wonderful and nice intro. Thank you. What stands out now? We're in a tough labor market, especially for younger workers. I think there's this anticipation of how AI might affect work that is not like AI is being adopted that quickly at large organizations, especially, but they're still pulling back in anticipation of these roles changing. And I think that disproportionately affects young workers because young workers are really a more uncertain bet. They're a more high-variance bet for organizations, whereas older, more experienced workers are kind of safer. I kind of know what you're getting. There's a faster time to ramp up. So the good news for older workers is that the impact on AI is very small on that demographic, if at all. There was a study that we did at Ravelia Labs, also a study done by Stanford economists, Eric Brynjolson, and his co-authors on the effect of sort of AI exposure on labor demand, on job posting volume. Big reduction in demand for highly exposed roles of entry-level workers, and basically nothing for highly exposed roles for experienced workers. That's part of the good news. I wouldn't be too comfortable just yet, because I think, like I mentioned before, what we're seeing right now is anticipatory. Companies are anticipating how AI might affect their work processes and the composition of work in their organizations, but that's different from how AI adoption will actually manifest in an organization when adoption increases. So I think for older workers, it's more about choosing the organization, choosing the subgroup in the organization, such that they are in more adaptable roles. I want to introduce this concept of adaptable versus procedural. There are some organizations that are super procedural where someone gets hired to be a specialist in very specific subtasks. Someone on the proverbial assembly line just doing the same thing day in, day out. And we know that technology generally, and AI especially, doesn't automate jobs wholesale. It automates components of jobs, it automates tasks. And how do we respond to that? We introduce new tasks, we find new things to do, we adapt. And if you're in a role where you're very specialized in something very narrow, it's hard to adapt. It is possible that AI can displace your tasks faster than you can introduce new ones. But if you are in more of an adaptable organization where work reconfigures all the time, then the effect of technology is actually not that big. I'll give you an example from my own company. We're like 70 people, data scientists, economists, and we're small and scrappy. So every day there's a new request from a client, there's a new initiative that goes on. Someone might leave their job or be out sick or on leave, or someone's not interested in doing this thing anymore. There's so many things that happen just every day, which requires us to kind of go back to the drawing board and say, okay, who's got the bandwidth to take this on? How should we kind of reconfigure who does what? And we're constantly reconfiguring the work that people do in their jobs. And that has nothing to do with technology. So if technology affects that in like a slow and secular way, that's actually just not a big deal for us. We're so used to adapting to the changing needs of an organization. What I would advise existing workers, experienced workers, is try to do more job crafting. Try to reconfigure your work, do things a little differently, kind of on the margins, just experimenting with it, see if your organization is open to it. And if they're not, I'd be nervous. I'd be looking to take more control over your career, maybe change jobs, maybe change groups within the company and associate yourself with leaders who are less top-down, this is what you do, and who are more partners in thinking about how work should get done.

SPEAKER_00 5:47

Hi, this is Ryan Kamori, founder and CEO of Saverlining, a company dedicated to acquiring teams to live more joyful, productive, and mentally fit lives. And we're a prod sponsor this episode of the HR Chat Show. We exist a bridge, a key activation gap for many out there with AyaCraft, a figurative mental health gap. Also teaching helpful skills that are not normally covered in weekly talk therapy sessions. Our goal is to help people everywhere become more aware of their own mental health and improve their mental fitness. Learn more about how we can help your employees at Saberlining.com.

Job Crafting And Leadership Fit

SPEAKER_03 6:28

Thank you. It's interesting advice to consider both personally, how we approach our development, our role, how we assess an organization and the development they're going to be able to offer us in the short term, but also the likelihood that our role or organization are going to continue to thrive in the longer term. And that pressure test, because we know that roles will need to evolve. Certain tasks will become automated. So all the better for leaning both for our learning and to pressure test organization.

SPEAKER_01 6:58

Yeah, for sure. It's beyond just a risk management approach where you can save yourself from potential automation. It's also going to make for a more interesting job. I think that jobs that can't transform are worse jobs. They have lower employee satisfaction, higher rates of attrition, and it feels micromanaging. Nobody really likes it. And I think in general, there should be more pressure for organizations to make jobs more dynamic. I think it helps them be more resilient. And it's good if workers kind of vote with their feet.

SPEAKER_03 7:29

Very good. And I'll note too that it will be a struggle for those who like to do the same thing in the same way when we look at change that will be forthcoming with what can be automated.

SPEAKER_01 7:41

I think that's right. I don't know how sympathetic I am. If you do like to do the same thing the same way every day, then that's kind of what computers are good at. So I would say like play to the strengths of humans. We're adaptive, we're very resilient, and we should be doing what we're what we're naturally good at rather than specializing in what computers are naturally good at.

Humans Shouldn’t Do What Computers Do

SPEAKER_03 8:02

That's great. Thank you. And I'll note as we look at later stages of the career, there's been traditionally, stereotypically, a lot of false assumptions around those of us further along in our career, whether we're resistant to change, when the reality is most of us in our generation have designed the change that we see forthcoming. And people can be rigid at any point of their life based on their personality, and also appreciate the call about how we need to lean into the flexibility that makes us uniquely human as well. Many traditional career models still carry an outdated narrative of decline at a certain point in our career. And it's far too early for my liking. I'd love to hear from you what the data is telling us about how this story is changing or how it needs to change.

Rethinking The Midlife Career Narrative

The Loyalty Tax And Exploring Versus Exploiting

SPEAKER_01 8:52

Yeah, it's a really interesting question. I mean, we do see that earnings do kind of drop after some peak when you're in prime earning status. I think there's a mixed story there. One part of that, and this is a little bit of conjecture, some of it is less about how much we're making in a certain job and more about just career choices and prioritizing flexibility. So when someone's more senior in their career, maybe they want to take off and spend more time with family and work on side projects or get more engaged with their community, there's more financial flexibility to do that. And that's just a choice that people make, which I think is great. That's good for them, and they should be able to do that. So I think the average decline is not so much of a concern because it includes a lot of just labor force participation choices. There is also a phenomenon that we call like the loyalty tax, where people do make more money when they move jobs. When people are younger, there's more incentive for them to move their jobs. There's this model in operations research and math called the one-armed bandit problem. Basically goes like this. Let's say you go into a casino and it's got a bunch of slot machines, and each slot machine has a different payoff, and you're in there for some fixed interval of time. What do you do? And basically the answer is that you spend a chunk of your time exploring and trying out different things, gathering data on the payoffs, and then the rest of your time you spend exploiting. So where you basically play the machines that have the best payoff. And that shows up in life all the time, in some fixed interval of someone's career or where someone lives and trying out restaurants or whatever it is. The optimal strategy is to explore early on, gather information, develop. And optimal strategy as you're further on is to exploit what you like and just keep doing the comfortable thing. There's some comfort there. We have a lot of information on what works for us, but I think that does carry some risk because you don't get to carry experiences from other firms. And taking on that kind of discomfort of hopping around has a financial advantage. So I don't know if it's necessarily the wrong strategy to be staying put, but it's got some financial penalty to it. I think it's worth being clear-eyed. Um, that is a choice, and we are facing a trade-off.

SPEAKER_03 11:07

Thank you. What are you seeing in the data about how people in their 50s and 60s are navigating their careers? Those in their 50s who may feel their development has slowed, those in their 60s who may start to feel like they're being nudged towards the sidelines.

Coordination Beats Execution As AI Scales

SPEAKER_01 11:26

There's a few things going on. I mean, when we saw this in previous decades of people who were older kind of being more sidelined, pressured to retire, I think there were a few things going on. One is that they were paid more and for some amount of output, they were just more expensive. So even though it's technically illegal to discriminate based on age, it I think it would be naive to assume that it doesn't happen. But then there's also, I think there was a big effect of an advantage to being technologically native. So I think now we're in a different paradigm because digital tools 10, 20 years ago had a steep learning curve. Digital tools now are so easy to learn. Generative AI is just writing text. That is something everyone can do. Everyone can do well. So I don't think there's a technological disadvantage. I would almost argue that they're at a technological advantage. I mentioned earlier this issue of coordination versus execution. So I want to go deep into that. I can think of jobs as bundles of tasks. You know, you do a collection of things. But as a worker, you're not just hired to execute on tasks. You're also hired to coordinate between those tasks. You're also deciding what comes first, what comes second, what does it interact with, who needs to be in the loop, and you're playing the role of an orchestrator, not just among yourself, but also among others. I think as generative AI becomes more powerful, it will become more powerful in delivering on tasks, in executing things that humans were traditionally better at. And the relative returns for workers are going to be on who's better at orchestration. And so I think that typically is older, more experienced workers. They are very often connectors in the organization. They're used to managing teams. And I could see a world where even entry-level workers kind of become middle managers in a way because they're managing AI systems and vendors and stuff like that that all kind of have to fit together. I think that skill is more prevalent among older workers. I think that's part of the story of why we're seeing younger workers being affected more than older workers by AI. I'm a little bit more optimistic in this latest wave of technology for older workers.

SPEAKER_03 13:45

Thank you. From your research, what factors most influence the decisions that people are making at later stages, whether it's to stay, shift roles, reinvent themselves?

What Older Workers Value Most

SPEAKER_01 13:57

It's a good question. I looked at this years ago using things like Glassdoor reviews to look at how much people value different facets of their job by generation, by age. I don't have the latest numbers on hand, but I remember more of a priority for work-life balance. And there's more of a priority for business outlook and management, sort of approval of management. Whereas younger workers were understandably more sensitive to promotion and pay and technology and things like that. I think that makes intuitive sense that as people get more senior, they're more engaged with the business strategy. They're more engaged with management, they have the ability to judge these things if they're headed in the wrong direction. That's probably quite a bit more frustrating for someone who knows the deal.

SPEAKER_03 14:42

Thank you. How are employers responding to the shifts? We're seeing research that people are staying in the workforce longer, even when there's sometimes societal or organizational cues to shift gears.

Risk-Off Employers And Lower Mobility

SPEAKER_01 14:58

I think it's a little hard to disentangle what is a secular trend over the span of decades and what's something that we're seeing in this short-term economic cycle. But one pattern that we're seeing that I think is attributable to the short-term economic cycle is that employers are much more conservative than they've been. We are in a very risk-off environment. And that could be uncertainty about technology. Everyone knows that generative AI is very cool, demos well, is probably going to change work processes, but we don't know how. And there's a lot of policy uncertainty. Tariffs are up and down, immigration restrictions are all over the place. There's a lot of change happening aside from just the technology of AI. AI companies are getting a lot of investment. All the kind of risk capital is going to AI companies. The financing environment is becoming much more risk-off. Private equity firms are focused a lot more on profitability than on growth. And that tends to create incentives for companies to focus on the short term. And I think that benefits older workers. Younger workers are seen as a better, cheaper investment for the long term. And older workers are seen as a better investment for the short term. They can deliver. So one thing we're definitely seeing is a dramatic fall in the rates of attrition. People are not leaving firms very quickly. We're seeing layoffs in some sectors of the economy, but for the most part, there's not a lot of mobility in between companies. Hiring rates are lower, attrition rates are lower. They're really at lows. So I think we're seeing a lot of older workers stay put and companies wanting to hang on to the workers they have and reducing hires for new workers.

Remote Work And Age Dynamics

SPEAKER_03 16:42

Thank you, Ben. It's an important call-out about the current state of the economy and employers' choices at this moment in time as well. And with stability, we see a lot less movement. We see a lot less risks, and we're approached a lot less of the round growth initiatives and much more about reorganizations. And we used to see just bifurcation where yes, there's interest in AI, but apart from that, the strategic work is really much more around looking inwards and risk mitigation containment.

SPEAKER_01 17:14

Yeah. There's one other phenomenon that I want to touch upon that I think also affects the relative trade-off between older workers and younger workers. And that's really around remote work. Older workers and adults in general are much more likely to want to work from the suburbs and not be in cities that are, for the most part, organized around young people. And that is something that a lot more companies are okay with now than they were six years ago when COVID was. But if you're young and single and dating, you have to be in a city. That's just how it goes. And if you're in a city, you're perfectly happy to work from an office. But for older workers, they're more okay being remote, and companies are more okay with that.

SPEAKER_03 17:52

Thank you. Do you see that impacting people's choices around career, limiting, expanding opportunities or thoughts on how that can be navigated well?

Demographics, Health, And Longer Careers

SPEAKER_01 18:03

Yeah, I think so. I mean, even just anecdotally from my own company, we hire remotely, we're a distributed team. And I think we we have an easier time recruiting more experienced workers because they have a preference for being remote. So if there's someone working out of Tampa or something, then great, they're part of the labor pool. But for companies that hire more younger workers, if you're working at like JP Morgan, they want everyone in the office, and that really affects their labor demand.

SPEAKER_03 18:28

If I can ask you to share some conjecture with the demographic changes around lifespan, talent shortages, technology reshaping work, what does the data suggest about how later career workers can or will contribute to the future economy?

SPEAKER_01 18:45

I think it depends on so many things. So one big question is what is happening to people's wealth? If retirement accounts are doing well, then people can retire earlier. Another big factor is health, of course. People are getting healthier over time. Life expectancy is increasing. Now with this recent wave of like obesity drugs, that's affecting things too. Then there's also fertility. Fertility's been falling for a long time, but there's less pressure to move in with someone's family and take care of grandchildren and things like that. That historically was a big factor for older workers leaving the labor force. They want to go spend time with their grandkids. Now there are less grandkids to go spend time with, and those grandkids are being born later and later. That's a big factor that I think doesn't get discussed enough. I think for those reasons, we have older people staying in the workforce. In terms of like how to contribute after leaving the workforce, that's probably more negative than positive, sadly. Of course, there's opportunities to be on boards and volunteer and whatever. But for the most part, when people leave the workforce, they're less likely to be contributing to society through informal channels.

Debunking The Lump Of Labor Fallacy

SPEAKER_03 19:52

As we think about the next decade of work, what gives you optimism about how data and insight can help redefine what the later stages of someone's career really means?

Job Architecture: Ben’s New Book

SPEAKER_01 20:04

The concept of older workers holding up opportunities for younger workers is absolutely crazy. I mean, this is also a concept that gets thrown around when people talk about immigration. The narrative is based on this assumption there's some finite number of jobs and there's some finite number of slots that someone could take. And, you know, when someone takes a job, they're crowding out opportunity for someone new to take that job. And that's something that in economics we call the lump of labor fallacy, that there's just some labor demand that's just given from the heavens. And that's really not right. I mean, people, workers, any people are both on the supply of labor side. Of course, they supply labor, but they also demand labor. I'm hiring people all the time to try to increase the demand for workers. I know you guys hire people too or have been involved in that. Someone, skilled person in the workforce, is more likely to be creating new opportunities than to be taking opportunities from someone else. That's a complete fallacy in that I think people just don't understand the economics of labor that well. In terms of how this is changing, whether there will be pressure for retirement ages to increase and people to spend more time in the workforce as they age, I'm a little bit more mixed on it. I mean, I think all the pressures that we thought of over the course of this conversation point to there being probably increasing value to experience an age relative to where we were even five years ago. So I think that's making me feel more optimistic. The thing that makes me feel more pessimistic about it is that there's a lot of political friction to actually formally changing retirement ages. And I could see that not changing because it's hard to change things like that. I wouldn't hold my breath for major policy change.

SPEAKER_03 21:48

I'm hopeful that as employers use informed data around their own workforce, hopefully increase the amount of conversations they're having, that they rely less on a Assumptions about what somebody's interests are based on their known age, and that we can become much more adaptable within organizations.

SPEAKER_01 22:09

I hope so.

SPEAKER_03 22:11

Then I understand you have a book launching. Congratulations. That's very exciting. What can you share with us?

Where To Follow Ben And Closing

SPEAKER_01 22:18

Yeah. It's called job architecture. The subtitle is Building a Language for Workforce Intelligence. The book is really about taxonomies. It's about how we categorize people into jobs and how we categorize people into skill groups and work activities. So it's really creating structure around what people do and how they do it. Being in people analytics for so many years, you start to realize that the success and failure of projects is really driven by what is your unit of analysis. Ultimately, you're analyzing groups of people. And the way people are grouped is based on very outdated, manually created structures that are just a total mess. I think bringing clarity to that is really necessary for any part of analytical HR. It's foundational. Another way of thinking about it is that, you know, sometimes I like to compare labor markets to capital markets. In economics, we talk about economic output being a function of two things, labor, labor, and capital. Roughly two-thirds of economic output is attributable to labor, one third to capital. Labor markets are twice as big as capital markets, but there is a whole science of allocating capital, and that's finance. And there are so many foundational pieces to make that work well. We have millions and millions of people whose job it is is to categorize information for the purposes of financial analysis. They're called accountants. And we don't have millions of people who categorize labor market information, but it really needs to be done. That's the building blocks of all analysis. In the book, I try to outline why it's really important, how to think about what are jobs, what are skills, what are work activities, how do they all fit together, how do we even have the right framework to think about the right job architecture? And then how do we do it? We can't hire armies of analysts. So how do we do it using the technology that exists today? How do we use LLMs to create similarities between groups of people, how to cluster them, how to label them? It's actually not that hard with today's technology. And then finally, it's about why it matters and kind of a vision for the society we could have if labor markets are as efficient and scientific as capital markets.

SPEAKER_03 24:17

Thank you. Well, I'm looking forward to enjoying your book, to learning from your work. Uh, and immediately I'm like, oh, can we upload that to our LLM? Right? Apply what's uh what's in your brain and what you've kindly shared with the world.

SPEAKER_01 24:33

Yeah, hey, go for it.

SPEAKER_03 24:34

Ben, thank you for your time and your insights. What's the best way for someone to follow your and Ravellio lab's work?

SPEAKER_01 24:44

Yeah, thank you. LinkedIn, very active on LinkedIn. I post insights about the labor market all the time, every week. That's probably the biggest. I got a book out um in January. So check that out. It's called Job Architecture, Building a Language for Workforce Intelligence. Those are the two that kind of come to mind. Be in touch, connect, follow, all that.

SPEAKER_03 25:04

Thank you. I'm grateful to you. Appreciate your time. I know you're very busy, so I'm I'm grateful for you making the time for this. Yeah, likewise. I hope you found this conversation with Dr. Ben Sweg thought-provoking with his insights on both the labor market and how work is organized. I'd love to hear what stood out for you in today's episode. Anything that surprised you or shifted or even validated your thinking. I'm Pauline James, and this is the next chapter. Connect with me on LinkedIn or through our Anchor HR community. Your reflections are always welcome. Thanks so much for listening, and I hope you'll join me next time as we continue exploring what the next chapter of work and life can look like.

SPEAKER_02 25:45

Thanks for listening to the HR Chat Show. If you enjoyed this episode, why not subscribe and listen to some of the hundreds of episodes published by HR Gazette? And remember, for what's new in the world of work, subscribe to the show, follow us on social media, and visit hrgazette.com.