Although top line unemployment numbers are still low, they don’t tell the whole story. Thousands of federal workers who have either been laid off or resigned are omitted from the official count; there’s a growing mismatch between the skills that employers want and the ones that job seekers actually have; and economic anxiety over Trump’s latest round of tariffs is sure to make companies at press pause on their workforce plans. 

It’s not just workers that are struggling, however. Recruiters are overwhelmed with applications, trying to figure out how best to utilize AI, and under huge pressure to sign on top talent. Daniel Chait, CEO and co-founder of Greenhouse, understands the stakes that both applicants and recruiters are facing—the hiring platform filled 2 million jobs in 2024 alone.  

Chait spoke with Fortune about the latest trends in hiring, how AI has complicated the job search for applicants and recruiters alike, and how to avoid a ghost job. 

This interview has been edited and condensed for clarity. 

Fortune: What are some of the major hiring trends you’re seeing in 2025 so far? 

Daniel Chait: To zoom out a bit, there’s not a job market with one thing happening. There’s a lot of different aspects to it and each of them have very different characteristics.

If you have friends in D.C. [in federal government], like I do, it’s an apocalypse there. If you look at certain kinds of frontline worker roles, in healthcare, nursing, or hospitality, there are tremendous worker shortages that have persisted for years. So there’s still a lot of power in the hands of candidates and applicants there. 

If you look at white collar tech jobs, though, you see this really strange situation where, on the one hand, there’s really low unemployment and the era of mass layoffs in tech is a little bit past us. But at the same time, it’s getting harder to get a job and people really aren’t moving. A lot of people feel frozen in place and AI is magnifying all the noise. Job seekers are using tools to apply to hundreds and hundreds of jobs automatically. So employers are seeing a deluge of resumes when they open up a role. It’s the sifting through that is really the big issue.

AI is changing the application process as we know it. How is it currently being utilized? 

Job seekers are using an AI agent to apply on their behalf. You pay $29 a month, you get your AI agent. It reads your resume, it goes and finds jobs for you, it writes cover letters for you, it customizes your resume and applies to hundreds of jobs at a time. So, on the one hand, it’s getting easier and easier and easier to apply for a job. What it’s not doing is making it easier to get a job.

When you look at what’s happening from the perspective of the employer, they are seeing their funnels just fill up with more and more resumes. But it’s not necessarily more quality applicants that should get that job. It’s just more noise. So what they’re [thinking] is, “Can I get some AI on my end to sift through all that and rank those resumes so I know the 10 or 20 or 50 I should read.” It’s just this mad arms race that’s getting worse, not better.

Another big issue is fraud. You’re seeing job seekers using ringers: someone to take the interview for them. Then someone shows up for the job that isn’t the person that took the interview. Famously, applicants are using AI tools up on the screen. So if I’m in an interview with you on Zoom and you’re asking me a question, I’ve got a tool next to me as a job seeker listening to the question and giving me the answers, and I can just read the answers back to you. And you don’t know that I’m using AI to fake my way through the test. 

There’s a bunch of that stuff happening in job seeker land that’s causing consternation in companies. And they want to know, “Am I really assessing the person I think I am?” 

What are some potential solutions to minimize risks like fraud with AI?

A big one is transparency. We actually published a policy for people interviewing for jobs at our company around how they can use AI. I don’t think the answer is “don’t use it.” The truth of the matter is, a lot of the jobs that people do at the company are using AI. So I want to know that people that may join this company are good at using this technology. At the same time, what I don’t want them to do is cheat. If I ask you a question about you, I want to understand you. I don’t want to understand what ChatGPT thinks about you. 

We publish guidelines that say, “Here’s how we expect you to use and not to use AI” and on the other end, “Here’s how we’re going to use AI to assess you.” There is a real discussion that needs to happen between job seekers and employers about how this stuff is meant to work in the process. 

What is the role of a recruiter like today?

It’s changed a lot. Our typical recruiter in Greenhouse is handling something like 20% more applications than they were a year ago. That’s been a trend for a while. There’s over 400 applications at a given time in any given recruiter inbox—and that’s an average. The outliers are way higher than that. And so it’s managing these bigger and bigger pipelines. They need to be more technology oriented. They need to be a lot more efficient with their time. 

The other big change that’s happening is what we used to think of as sourcing. Going back a couple decades, if you talked to a recruiter, the value they had was their Rolodex, who they knew. I remember talking to recruiters in 1999 and they knew, for example, the 50 best programmers in lower Manhattan. If you wanted to hire a programmer in that neighborhood, you called this person, and they knew them. And then LinkedIn came up, and now everyone’s got a Rolodex with 850 million names in it. 

It’s now rather how quickly can they sift through those 850 million names and get to the ones you care about and to the job of a source. It’s less about knowing people and more about using technology to search efficiently, having good judgment to surface those people to the hiring manager. 

The jobs are changing very, very quickly in recruiting operations: managing big technology stacks, grappling with AI, dealing with fraud and risk. Recruiting operations is very different and much more sophisticated than it was even a few years ago. 

There are a huge amount of ghost jobs out there. What needs to be done to prevent job seekers from wasting their time? 

Companies are going to continue to have some amount of jobs that, for whatever reason, are listed on their website  but aren’t as a high priority or aren’t as active. Our data shows that it’s about one in five. So it’s, it’s a real dynamic, but it’s not most jobs. So when four out of five jobs are not ghost jobs, and people still are applying to them and not hearing back, I think the answer is less about the ghost jobs.

It’s just like there’s thousands and thousands of applicants in every queue. I give the analogy that it’s like a deli. When you go to the deli, you take a number, right? And you’ll sit there and wait. But if you went to the deli and you took a number and it said you were number 3,000 in line, you would leave. Yeah. But what’s happening is that [when] people are applying to a job—they just got in line at the deli and what they don’t know is they’re number 3,000. They stand there for a few days and they’re like, “I’m not getting my sandwich with this place. This deli stinks.” No, you’re just way behind the line and you don’t know. 

Whether it’s a ghost job or whether it’s just the fact that there are a lot of applicants [for] any given job today, I think giving the job seeker a lot more transparency about [whether] we’ve looked at your resume or we haven’t, here’s how fast the line is moving, [will] people a clearer sense of what’s actually happening. 

What do you think the job application process will look like in the next five years? 

Everyone’s going to have a full-time job-searching AI bot on their own behalf, scouring the web and applying to all kinds of jobs all the time. And so as the number [of applications] goes to infinity, the value [of each application] is going to go to zero. At the same time, the human element is going to stand out even more. The number of people that are gonna take the time and do the thinking to actually find the [recruiter] and say something compelling to them—that’s a real skill. 

And so I think it’s less about the volume, it’s more about the specificity and human energy that you put into it.