What the jobs report isn’t telling you

Note: Transcripts are automatically generated from episode audio, and are not fully corrected for spelling, grammar, and formatting.

HAYDEN BROWN: The future of work is arriving, and it’s arriving on our shores right now, right here, whether we want it to or not. The urgency and the need for all of us to become fluent in using AI for our jobs is something none of us can escape. You can’t learn to ride a bike in a seminar, and I think the same is true for AI. Until you’re using these tools day in and day out, really integrating them and pushing the boundaries of what they can do, that’s when people become extremely capable. Will humans win, or will machines win? I think that is distracting us from, in some cases, putting our collective societal energy toward building a future where everyone wins.

BOB SAFIAN: That’s Hayden Brown, CEO of Upwork, the global tech platform for freelance workers. As AI radically resets expectations across the workforce, Upwork has a critical vantage point on how the job market is actually evolving. I wanted to talk to Hayden about the on-the-ground reality beyond government job numbers and the dreamy predictions of AI cheerleaders. Hayden is forthright about the pressures facing organizations and their teams, including her own decision to lay off Upwork staffers earlier this year. She shares which jobs are most in demand, the good and bad of AI agents, and where she sees the future of work moving. Plus, she talks about the challenges of leading when even a one-year planning horizon is too far, revealing her specific practices for resetting strategy on the fly. So let’s get to it. I’m Bob Safian, and this is Rapid Response.

[THEME MUSIC]

I’m here with Hayden Brown, president and CEO of Upwork. Hayden, great to see you.

BROWN: Great to see you, Bob. Thanks for having me.

SAFIAN: You were last a guest on the show when we were talking geopolitics, Russia, Ukraine. It was pre-Great Resignation, pre-return-to-office controversies, pre-Gaza and Iran. Now, of course, AI anxiety dominates workforce discussions. It’s like we’ve been through decades of change all piled together. How are you doing? It’s been a challenging time to be a CEO.

BROWN: It definitely feels like there’s a lot happening. I have a running joke with some friends who run other companies. We all say, “What’s going to be the black swan event of this quarter?” Because it just feels like there is so much happening in the ecosystem, and things are moving faster and faster. But I think that also is very exciting. It’s a dynamic time.

SAFIAN: I was going to say, a black swan every quarter, or maybe a black swan every week. You don’t even know if it’s a black swan until it’s already moved on, right?

BROWN: Exactly.

Copy LinkMaking sense of the monthly jobs report

SAFIAN: You’re on the front lines of changing work habits, and I find myself wondering a lot if traditional assessments capture the realities of the U.S. labor market in 2026. The most recent monthly jobs report showed 23,000 jobs lost, but at the same time, unemployment numbers went down also. How do you make sense of this data? How much do you focus on it? Does it reflect what you see on Upwork?

BROWN: Yeah. There are a lot of competing dynamics. It certainly doesn’t tell the whole story. We do see, kind of consistent with the jobs report data, that the labor market is very subdued. It has been for several quarters. It continues to be sluggish. I would say we particularly hear about macroeconomic and labor slowdown from our very small-business customers. Those are the guys who are out on the front lines of the economy without a lot of resourcing, without deep pockets and financial reserves, and I think they’re the ones who are feeling quite pinched right now. What they tell us is they have pulled back hiring across all channels.

SAFIAN: You guys recently released a “Future of Work Index.” Is that what encapsulates this? Are there other things in that index that business news headlines maybe aren’t capturing?

BROWN: Yeah. While the picture overall is of this kind of slow-moving market, there are pockets within it that are incredibly vibrant. We really see the growth in AI-related skills and AI-related jobs. That is the place where energy and money is flowing, and we see that on our platform, with our own AI category growing more than 22% year over year. All of the data is telling us that the demand from both small businesses all the way up through large enterprises for talent who really know and can help them navigate this new world of work, and bring in these relevant skill sets and knowledge of how to use AI tools, is really where the energy is going. And the dynamics are moving from full-time toward fractional work, which is a really fascinating trend. Our most recent survey showed a 10-point increase in freelancing in the U.S. among knowledge workers. It’s at 38% of the population today. It was at 28% just a year ago. So there has been a huge step up in both workers moving into the freelance economy for a variety of reasons, and businesses saying, “We want to find these workers who are out there doing contingent work because we don’t necessarily need the same number of full-time employees we needed before. We can approach this work differently, and also, we can’t access these folks maybe through full-time employment channels.”

SAFIAN: There’s so much talk about AI potentially taking jobs, and there are some layoffs that companies attribute to AI. Yet you have other CEOs claim that AI is allowing them to add staffing because each AI-enabled person is more productive. It sounds like both fit in what you’re seeing, but it seems early to know what AI’s impact is going to be. Or is there something clearer that you see through all of this?

BROWN: I think it’s 60% of jobs that exist today did not exist in 1940, and I think that’s helpful context for us as we look at what’s happening with AI. Technology and other forces have destroyed and created jobs again and again through so many cycles. This is kind of a version of that, but happening bigger and faster, more destructive on the negatives, but also more creative on the positives. It’s a very binary picture. There’s the camp of AI is going to take all the jobs, and humans and the companies that serve humans and do human solutions are doomed, and then there’s the other side. The reality has to be somewhere in the middle. We do not see mass job destruction as the future. There is really, I think, a balance to be struck. Again, there’s creation and destruction, but at the end of the day, it’s not all destruction. It actually is a lot of new jobs and roles and businesses that are being born because of this exciting technology.

Copy LinkWhat AI is changing inside companies now

SAFIAN: You announced in May that you had to cut a quarter of Upwork’s corporate staff yourselves. You published a pretty frank statement about it, owning the decision, and you talked about AI, but didn’t blame AI. Why was that necessary?

BROWN: Yeah. I made that decision earlier this year because we really did see that, with the market conditions changing, this was the right time for us to expand the profitability of the business, which was already part of our long-term goals. At the same time, Bob, we’ve seen that because work is changing, our own work needs to change. What we did for that was not really just automating and replacing work or workers with AI. Instead, we looked across the entire business and asked ourselves in every team and every department, what would this look like if we really were doing this function from the ground up today using the best technology and tools that are available? We also looked at newer operating organizational models where we could flatten the teams, get more leverage for managers with broader spans of control, fewer layers in the organization, some of these things that are kind of evergreen opportunities to refine how we operate.

SAFIAN: There are contentions, I guess, that some CEOs do AI washing. They ascribe to AI the reason why they have to make changes, as opposed to owning that maybe they didn’t have the right strategy in the first place. But by blaming it on AI, they get credit for that and their stock price goes up. It’s like it’s a positive. And I wonder whether, as you were thinking about this, you were like, “Well, do I lean into that part of the messaging?” Because maybe that helps the story.

BROWN: It was so important to me that our messaging around this be really true and authentic because, frankly, of course AI is helping us, and it is a contributor to our efficiency and our productivity. But that is not the biggest reason, A, that we made the changes, nor the biggest benefit that we were able to get from driving these changes. We’re actually seeing the opposite trend on our platform in the sense that the AI washing you’re talking about has been happening, but also a lot of folks are now reconsidering and coming back because they’re seeing that the humans they let go, they actually still need, and the AI can’t do all of the things they hoped it could do. A recent survey we did showed that 23% of our own clients have actually already moved work back to humans from AI, or are about to do so. So I think this pendulum will keep swinging. When we’ve actually done testing of clients trying to use agents directly to deliver work products through our platform, the failure rate of agents is incredibly high. They almost never succeed in delivering even fairly simple tasks from clients end to end. But when you just add a little bit of human expertise and a few back-and-forths with an expert, suddenly the success rate goes up more than 70%. We have to sharpen our intuition around when to trust AI, and I think increasingly people are realizing, I’m not going to take this at face value. I want to couple that with a valued human expert who can bring a lot more to that equation.

SAFIAN: AI is directly impacting things that happen on your platform. There’s been a rise in AI agents posting job ads, machines essentially trying to hire humans. Upwork even has, I think, a specific service for clients to allow agents to access your freelancer database. It could sound a little spooky. Did it feel that way to you when it first started?

BROWN: I mean, it really feels like the future of work is arriving, and it’s arriving on our shores right now, right here, whether we want it to or not. That was one of the reasons we launched this what’s called the MCP server, but really it’s a capability to let agents authenticate with Upwork so we know who they are. We know if they’re good actors or bad actors. And then, based on that authentication, they can go forward, help post jobs, make hires, manage projects, et cetera. Now, we’re definitely doing this because this is where the world is going already. It’s not something that we just came up with out of the box. At the same time, we’re seeing a lot of clients who are working inside these AI tools, Claude or ChatGPT, and they’re getting stuck. They have a deliverable that didn’t meet expectations, or they need a human expert to finish it. With our MCP, now they can directly connect to Upwork, share that deliverable with a freelancer immediately, and get the on-demand help they need from a human expert in real time.

Which is a huge advantage because now everyone expects everything in real time. Everything’s moving so fast. AI has changed expectations, and now we too can meet that expectation around truly real-time human support inside any of these AI tools where people need additional resourcing.

Copy LinkWhy human proof matters more than AI polish

SAFIAN: Upwork has its own agent, Uma, that assists users writing job descriptions and prepping interviews and things like that. For a platform conceived to give humans more freedom, there’s some irony in using AI to prepare a job application so another AI can review it, right? Are we entering a labor market where the best job applicant is whoever is best at prompting the hiring machine? Or is it more like online dating apps, where algorithms are better at matching us than we are on our own?

BROWN: These are the big questions, and they’re the right ones, because we are seeing spam job proposals and applications. We really just want to put people’s own interests and needs on steroids, and then have that expressed through the hiring process and the recruitment process that we’re supporting them with, including filtering out all of the AI slop and noise that is creeping in now that these tools are so widespread.

SAFIAN: Right. I guess if I’m a job applicant, I can create a lot of different versions of my application for a lot of different jobs that may or may not be right for me, and I may not even know what I’m doing. I guess the same thing can happen on the other side about posting for jobs, where if we can find a needle in a haystack, we’ll do it, but otherwise we’re just sort of testing the water.

BROWN: Yeah, and I think that’s one of the main values that we see being so enduring, even in the AI era, for Upwork specifically and our customers. The problem we’ve always solved for them is giving them real, tangible, accurate data around who can do what. It’s real data from our platform, observing people doing work and delivering for customers, with dollars attached. So this is the most verified work history data, really, in the world, and it’s at scale across 18 million freelancers who were active in the last year. This type of data is what gives customers confidence in making that hire, not noise that was created by an AI agent applying for them, not these empty promises or claims about their expertise. It’s actually stripping us back to what the data says in terms of who’s good at what?

And frankly, advancing this meritocratic idea that it shouldn’t be about who has the best agent or who sounds the shiniest on paper, but really who has the data and the experience to demonstrate that they can do the work that’s required.

SAFIAN: That in-the-trenches data Hayden cites illustrates how human-in-the-loop is actually unfolding in an AI world: 23% of businesses moving work back to humans from AI. So how does Upwork engage with the big AI companies, including open-source players in China, and what does it mean that freelancers with AI skills earn 34% more? We’ll talk about that and more after the break. Stay with us.

[AD BREAK]

Before the break, Upwork’s Hayden Brown talked about what’s really going on in the job market as AI shifts expectations and needs. Now she talks about dealing with the big AI companies, how to actually build AI skills effectively, and what Hayden calls her crystal ball, plus her specific strategies for leading in a world where, as she puts it, even a one-year planning horizon is too far out. Let’s jump back in.

Do you have a philosophy about how you engage with the big AI players?

BROWN: We’re at a moment where everything is in flux. We have this growth of these new channels where customers are suddenly doing work inside of ChatGPT and Claude and all of these third-party tools. That is now a new origination point for businesses like ours to be referenced, to be advertising, to be showing up.

We also know with 900-plus million users on ChatGPT and Anthropic, growing so massively quarter by quarter, I think we can see that these are the new destinations for work. The momentum there is building and snowballing in such a way that I think we’ve taken the approach of building for these tools and this future that we see already appearing, even though there is going to be some ongoing volatility as things mature and play out.

Copy LinkHow Upwork weighs AI cost against control

SAFIAN: Do you have a differentiation between places like OpenAI and Anthropic versus the open models from China? You do have a global business. Do you let agents built on those models be on Upwork? Do you have to treat them differently?

BROWN: We look at everything in our authentication layer and do make some judgments about who is allowed and not allowed to operate in our ecosystem, and certainly those rules will continue to change as new information is available about safety, security, and other aspects of what these actors can do. So I think our approach today is to create the system, the framework, and the controls. Our customers really have told us they want to be using these things. They want to benefit on both sides, but they need to do so in a place that’s not the Wild Wild West.

SAFIAN: I ask because some CEOs I’ve talked to recently have been talking about how some of the models from China are just so much cheaper.

BROWN: They are.

SAFIAN: And even though they’re not maybe quite as robust, they’re robust enough.

BROWN: Absolutely. And I think a lot of those open-source models can be used safely and securely with the right controls. It’s not a binary of use Claude or good luck to you and there’s nothing you can do. We personally, in our business, have taken a gateway approach where we are constantly testing the performance of a variety of different models, including open-source models. And now that token costs have been going up so significantly from the major providers here in the U.S., there are definitely use cases where we know we can get 10x savings or more by using open-source models and doing it safely inside of our own environments. I think this is also a place where, for us as a marketplace, creating more of the ecosystem where people can see the pros, the cons, and the costs of different combinations of AI models or agents, with or without human supervision, is a fascinating extension of our own business because we’ve always provided customers with the value of choice and selection across a huge supplier base to get something done, and they get to choose, “Do I want to pay more for speed and quality, or am I happy to go a little bit slower and maybe have some savings?” These are the decisions businesses always need to make, and now with these AI tools, they can kind of get the best of both worlds.

Copy LinkWhy employees are becoming freelancers

SAFIAN: Your emphasis on freelance work and fractional work, as you talk about that, there’s been a sort of romantic sentiment behind it: the promise of freedom and making your own schedule. In today’s economy, how much of that is a choice versus a necessity, a symptom in some ways of companies shifting risk onto individuals?

BROWN: Absolutely both. Fifty-eight percent of current full-time employees are actually thinking about becoming freelancers for a lot of the reasons you mentioned around choice and control and deciding when, where, and for whom they work. We also hear really loud and clear from freelancers that the economics of freelancing have become much more appealing over the last few years. Before the 2008-09 Great Recession, people at that time, I think, felt they were very secure in a full-time job.

Then that recession happened, and they realized, “Wait a minute, I’m actually more secure if I have a portfolio of three to five clients on Upwork. They keep coming back to me for repeat work. Maybe I’m also doing a podcast or a book or a creator-economy thing on the side.” And we really have seen this trend toward these portfolio careers. Gen Z is the biggest one. Forty-two percent of Gen Z are freelancing. It’s actually them saying, “Look, I feel more comfortable that I’m not exposed and that I can really both learn and use and build the skills I want and also have income that is predictable and in my control, versus at the whim of an employer.”

SAFIAN: It’s just a different conception of what a career is for younger folks, right?

BROWN: Absolutely. And it also reflects, if you think about this AI moment, when everyone is trying to prove that they have these skills, or if they don’t, learn them and then prove, “Hey, I’ve gotten these skills now. I’m an AI-relevant employee or worker,” freelancing is an amazing way to do that because they’re getting on-the-job experience, doing real work, proving their skills. And if they’re on a platform like ours, that comes with a verified work history, all of the marketing around that, that freelancers can do to say, “Hey, now I can earn more and do more.” And we do see them earning more. Freelancers who have AI skills are earning 34% more an hour than folks who don’t have AI skills. And I think that wage gap really speaks to the urgency and the demand from talent to move into a career where maybe in their full-time job, they don’t have the opportunity to use these tools, or they’re not learning them inside their employer. But here on our platform, they can really demonstrate and flex their muscles around AI and get paid for it.

SAFIAN: There’s a lot of burden on the individual to keep upgrading those skills as you go. When software can learn skills way more quickly than humans can, which is what we’re seeing with AI, then it does put more pressure on humans, I guess, whether you’re in a freelance economy or anywhere, to keep upgrading your own skills or refining your own skills.

BROWN: That is the reality. The urgency and the need for all of us, you, me, all of us, to become fluent in using AI for our jobs, that is something none of us can escape. But what we do see is that freelancers, because this actually pays them and puts food on the table, do it faster. They learn faster. They lean in harder. They move much more quickly to where the new technology is and learn its capabilities and apply those capabilities in real work settings. We see this huge advantage, actually, with the freelance talent pool that’s well ahead of full-time employee counterparts in terms of their adoption and facility with AI, and that’s another reason why businesses are turning more and more to the freelance economy. They’re looking in-house, and they’re realizing, “My team doesn’t have these skills, and it’s going to cost a lot of time and money to try to get my team up the learning curve.” Maybe I can short-circuit that by bringing in freelancers who can both add value and do work, but also upskill my team at the same time as a byproduct of sitting side by side with them.

SAFIAN: So it becomes a backdoor way to train my own team at the same time.

BROWN: Exactly, and not have to wait for those fruits to materialize before you start to actually see some real business benefit, because a freelancer can be adding value in hours.

Copy LinkHow to build AI skills

SAFIAN: Yeah, I mean, that is one of the big questions all of us have: How long does it take? What’s the best way to train people in AI skills?

BROWN: I was actually talking to one of our clients this morning, and he said, “You can’t learn to ride a bike in a seminar.” I think the same is true for AI. You could watch a YouTube tutorial and get more familiar, but at the end of the day, until you’re using these tools day in and day out, really integrating them and pushing the boundaries of what they can do, what the limitations are, why that is, and how to work around it, that’s when I think people become extremely capable. That’s one reason we’re seeing the rise of a new generalist called the AI orchestrator. This is becoming extremely popular on our platform because people are looking for someone who can orchestrate and use these AI tools, but who is not necessarily so deep in one domain. Instead, they have the facility to evolve, adopt new tools, and figure them out very quickly because we know the technology is moving so quickly.

SAFIAN: In what you do as a CEO, how are you using AI?

BROWN: I think AI, obviously, has core skills in data gathering, generation, and synthesis. Inside our environment, we’ve set up a very powerful tool we call Crystal Ball, and it has access to all of our internal data. This is wonderful for me because instead of having to ask one of my team members, “Hey, send me the most recent report on this,” or, “What happened on this day with this customer type?” I can directly query our database through natural language and get tons of insights around real-time business information, which we are then using in real time to make different decisions.

Copy LinkLeading with shorter plans in unstable times

SAFIAN: As I alluded to at the beginning, you’ve been CEO through this extraordinary period: the pandemic, the Ukraine invasion, which prompted you to shut down operations in Russia and Belarus, and a backlash against globalization, which also kind of threatens Upwork’s premise. One CEO talked to me a while back about building a cadence of change inside his organization to keep it agile. But now the cadence of change is coming from outside, and I wonder how that impacts how you make decisions and how you know what to focus on.

BROWN: We’ve worked to try to remove noise in the organization so that it’s very clear what small number of things we’re working on, and then, if we need to change one of them, we can communicate that really clearly once that decision is made. I think the more cruft there is in the organization around competing priorities and people moving in different directions, the harder it makes navigating these types of environments because it’s so unstable externally that you need to have locked-down stability and focus internally. We’ve also changed our planning horizons. I’d say in the past, I was a product manager for many years before becoming a CEO, and you would have a one-year roadmap. It would change and everything else, but it didn’t feel unreasonable to have some idea of what you wanted to be doing for a full year. Now, in product, if you have a one-year roadmap and you’re really executing the same thing in month 11 that you thought you would be when you started, something’s probably wrong because you will have learned so much more, and the technology will have become more capable. So we’ve shortened some of our planning cycles, still setting big, ambitious goals out in the future, but not being as committal to the steps required to get there. It’s kind of like, let’s take the first step and then reevaluate: What is that second step? Let’s not presume that from this vantage point today we know what that is.

SAFIAN: When you’re trying to focus and prioritize, do you have tools that you use to do that? I remember talking to Daniel Ek at Spotify at one point, and he would say each quarter he’d create a list of 10 things, but they were ranked. You couldn’t go to number two until you’d satisfied number one.

BROWN: We use ranked lists all the time, like Daniel. I think that’s a really important one. We also look at what our portfolio is: how much resource we’re putting into things that we have really high confidence will deliver in the near to medium term because it’s a road we’ve gone down before and we know it well, versus where we’re putting resourcing into things that are much more speculative or building for future states that we anticipate or can kind of see coming, but may not have arrived yet. I think it’s become even more important in this environment to be very clear and have internal alignment around how much of each of those activities you’re doing. Also, when you make changes, maybe something that was out in the future six months later is here. It’s arrived today. It has to move into that other part of the portfolio, and then you’ve got to be thinking, OK, where are we investing in that next horizon now that we’ve accomplished the thing that was out there in front of us?

SAFIAN: What’s at stake for the future of work right now?

BROWN: The binary debate that’s happening around whether humans win or machines win is distracting us from, in some cases, putting our collective societal energy toward building a future where everyone wins, where humans win with AI and are empowered by AI, but not at the loss of jobs, opportunities, and purpose that people have and will, I think, always need from doing meaningful work.

SAFIAN: Well, Hayden, this was great. Thanks so much for doing it.

BROWN: Thank you so much. Great to see you.