In his role as Managing Director at AMS, Jonathan Kestenbaum is continuing to build a rich career as an entrepreneur in the work tech space. A qualified attorney who has built and evaluated a plethora of work tech companies, Kestenbaum considers 2025 the most exciting time to work in this space. In this episode, he discusses how AI is helping us rethink the assembly line of knowledge work and the parallels between AI and the industrial revolution. We also discuss the impact of digital twins and AI administrative assistants with individuals and the law in mind, explore how to train staff on new technologies, and get into the higher-value roles recruiters will need to step into in response to the rise in accessible AI support. Join us for a deep dive into the future of staffing and the legal implications of AI, and the exciting possibilities that await us in this industry.
[0:01:14.4] David Folwell: Hello everyone, thank you for joining us for another episode of The Staffing Show. Today, I am super excited to be joined by Jonathan Kestenbaum, who is the managing director at AMS. Jonathan, really great to have you on the show.
[0:01:27.7] Jonathan Kestenbaum: Thanks for having me, excited to be here.
[0:01:29.4] David Folwell: Yeah, same, same. I’m excited about the conversation. We’re going to be jumping into kind of the future of staffing and talking deeply about AI, some of the legal applications. To kick things off, Jonathan, could you give a little background about who you are and how you got into the staffing industry?
[0:01:44.3] Jonathan Kestenbaum: Yeah. So, I am a licensed attorney by education, but I’ve been an entrepreneur in the work tech space my whole career. I started my career, I built a reverse auction system, which was essentially for tutors. So, students would ask for help, and tutors would bid against each other, ultimately helping students get help at the cheapest price. What I thought I was building at the time was an ed tech company.
I learned pretty quickly I was building a work tech company, helping tutors get jobs. A number of staffing firms on the East Coast were interested in leveraging the reverse auction for their contract staffing business to get higher margins. I ended up selling the company in 2012, and then I met Gene Holtzman, who is the CEO of Mitchell Martin, which is an IT healthcare staffing firm in the New York area.
And we cofounded an organization called Talent Tech Labs, which at the time was meant to be a venture fund to invest in work tech companies, but evolved pretty quickly into a research firm because in an effort to get companies to take us seriously as investors, we would write about trends in work tech, and then we get calls from heads of talent and CEOs and staffing firm saying, “What is this work tech stuff mean to me?”
And ultimately, built a research firm, which you know, we exited, and I’ve spent my career evaluating, you know, either building work tech companies as an entrepreneur, evaluating work tech companies for heads of talent, CEOs of staffing firms. I know every feature, functionality, revenue model that exists in this space, and I always joke that you know, they’re always the first, the only, the greatest. Meanwhile, you know, you get pretty cynical over time.
[0:03:22.2] David Folwell: You know, you’re the 7th first.
[0:03:24.1] Jonathan Kestenbaum: Yeah, exactly. But it’s, I’d say, the most exciting time to be in the space. I mean, I’ve been through the social revolution, the, you know, move from on-prem servers to SaaS based servers, the mobile shift, you know, this drive with AI is, I think, the most exciting.
[0:03:42.7] David Folwell: Yeah. It’s going to be transformational, it already is, and one of the things that you created that I waited for, looked at for years, was the talent tech ecosystem, that infographic. I remember the first, on that thing.
[0:03:53.8] Jonathan Kestenbaum: Yes.
[0:03:56.1] David Folwell: And like, you’ve probably invented categories within the space, over the years.
[0:04:02.9] Jonathan Kestenbaum: Absolutely, yeah, and those things have become budget categories, which is crazy.
[0:04:07.9] David Folwell: Yeah.
[0:04:08.8] Jonathan Kestenbaum: Yeah.
[0:04:10.1] David Folwell: What led you to that? I mean, for those of you that don’t know, there’s a talent tech ecosystem, I don’t know if that’s exactly what it’s called, but it’s an infographic. It shows you all of the HR tech and all of the different categories, and it’s grown massively over the years, and what the number of tools in that infographic, it’s wild.
[0:04:10.1] Jonathan Kestenbaum: I think what happened was what I kind of learned over time was the technology vendors would speak feature functionality, and like, you’d be on it, and depended on who you are on the call with, you know? But if you’re on a call with the founder, or even the, you know, someone on the technology team, like, they want to brag about, “This feature does this,” right?
But the buyers of these technologies, they talk business problems, and so we had to come up with a way to translate the noise between feature, functionality, and business problems, and so – and if you ask the vendors what problems they solve, they would be like, “We solve them all.” They’re either, “We’re do it now, or it’s on our roadmap, we’re going to do that for you, as soon as you sign up,” right?
So, this gave us an opportunity to kind of be like, “Well, what do you do today? What are the features that you have today? And if we were to look across all the companies that do that, well, can we roll it into a category that solves a business problem?” You know, maybe it’s referrals or –
[0:05:24.3] David Folwell: Yeah.
[0:05:24.3] Jonathan Kestenbaum: Maybe it’s video interviewing, and that’s probably what we did, and really, that marketing, it was really a marketing exercise. It was meant to get more companies to take us seriously as investors at the time.
[0:05:37.4] David Folwell: Yeah.
[0:05:39.0] Jonathan Kestenbaum: And you know, it really did accomplish very much that. Like, we got companies to tell us they wanted to be on the ecosystem, or death threats because you know, they thought they were in the wrong category, and we got, you know, heads of talent and CEOs of staffing firms saying, “Hey, which one’s the right one for me?”
[0:05:53.8] David Folwell: Yeah, that’s pretty incredible, that’s funny to think about all of the background into that. I didn’t realize it started from that perspective on your point on the way tech companies describe their products, versus the customers. I remember hearing about bidirectional SMS messaging as a, “That’s what we do.” You know, like if you ask their customers, they’re like, “Well, no, they just help me get a hold of talent.” It’s like –
[0:06:19.6] Jonathan Kestenbaum: No, I mean, like – and over the years, there’s been companies that, you know, said, “Oh, we’re not a CRM, we’re a TRM.” You know, a talent relationship management, instead of it can’t, and like, what I’ve learned is, you know, you’re better off driving into an existing budget category and going after existing spend within these organizations than plotting your own path and creating your own category.
It’s just too expensive. I mean, just think, like, in practice, the way this plays out is you go to a company and you’re like, “Hey, here’s what I do, and will you buy my technology?” And generally, the organization’s budget at the beginning of the budget period for a set of technology, and if you don’t fit into a category that they budgeted for –
[0:06:58.3] David Folwell: Yup.
[0:06:58.7] Jonathan Kestenbaum: You know, it’s very hard for them to find the cash, and when you come in with some made-up term of what you do, that you know, it’s not going to fit into an existing category.
[0:07:07.5] David Folwell: Completely fair, completely fair. So, kind of jumping in to where the market’s at today, and also we’re going to dig deep into AI, but why don’t we start with like, where do you see the kind of staffing HR tech market today, and then let’s jump into how tech’s going to impact that.
[0:07:23.6] Jonathan Kestenbaum: Yeah. So, I think we’re at this like, inflection point, and I used to do this, you know? And you’ll probably find it somewhere online. I used to do this talk on AI, and I used to talk about how we have, you know, with machine learning, the ability to do better predictive modeling and matching, but we really can’t do conversational regenerative tasks, and actually, now, with generative AI, we can do conversational and generative tasks.
And what that means is, you have now, for the first time, technologies that go beyond just pure workflow solutions to technologies that are actually able to do tasks that humans did, historically, and it’s creating a massive opportunity, but also it’s incredibly difficult to execute on. If you think about it, historically, if you bought a piece of technology, like a workflow tool to help you get efficiencies within your business, you would have to select the right tool.
You’d have to do change management and implement it correctly, and maybe, you’d get adoption, right? Now, you have to do those two things, but then this tool is constantly eating at more tasks that your people do, and so you have to constantly redesign the workforce that sits around it, and that’s hard. And so, you know, we’re going through this really interesting shift of likely, and I’m making up this stat, it’s just like a gut feel stat, this one.
So, something along the lines of 50% of the tasks that we do as folks within the staffing market are going to change. It doesn’t mean, by the way, like, I live in an abundant mindset, not a scarcity mindset. Like, my view probably is, yes, in the short term, it’s going to create less opportunity. Like, you know, long-term, it’s just going to give you more jobs. I can tell you, like, if you look at myself, I have 40% of what I do, probably, historically has changed.
I’m using AI a lot to support me and my job. But that doesn’t mean I have 40% more free time; I’m just busy doing more strategic things that I didn’t have time to do before.
[0:09:28.2] David Folwell: Yeah, I couldn’t agree more, and I think, in our first conversation, you talked about AI kind of almost commoditizing knowledge work. I’d love for you to talk a little bit more about that concept.
[0:09:39.4] Jonathan Kestenbaum: Yeah. Like, the way I tried it, and this actually goes into just the way my brain works, and though I even thought about creating categories around the Talent Tech Labs ecosystem, you know, you look for analogies, either historically or in other industries to see, like what worked, right? And if you think about it, if you look at the industrial revolution, which in the industrial revolution, you had machines.
And machines became the great equalizer of physical labor, and essentially, if you were strong before machines, you were like a hot commodity in a factory. Like, you would get a premium as a worker. Then, machines came in and, like, you know, they can get the pipsqueak from down the street to come and do the same labor that the strong guy or girl is doing before. So now, what we have is the great equalizer of knowledge work, right?
So, AI agents are going to kind of make us all smart to an extent, and obviously, how you use the tools is unique, and so there’s going to be folks that know how to use these tools more meaningfully and take advantage of that more meaningfully, and they’ll get more output, but broadly, we’re creating what I would call the assembly line of knowledge work, and we’re going to be reshuffling the task that we do alongside a machine.
So, if you look at an analogy of what happened in the industrial revolution, well, there was significant job loss at the beginning, and then there was, you know, significant abundance in the end, and more jobs created, and I think you’re going to see the same thing happen now, but even faster because the technology is evolving even faster. I mean, machines needed to be built, you know, with physical parts and welded together, and this is all zeroes and ones.
[0:11:19.0] David Folwell: It’s crazy to see how fast – I saw like the knowledge curve on NBC, it’s just a straight, straight high, it’s sick.
[0:11:25.3] Jonathan Kestenbaum: Really?
[0:11:26.5] David Folwell: We are moving at a pace that is amazing, and when you look at the roles within staffing, you say, and I agree with you. I think, to some people, like, “Oh, it’s overhyped, it’s too much.” And now, I think it might actually be underhyped. I mean, that’s a possibility that people may –
[0:11:41.4] Jonathan Kestenbaum: I believe it to be underhyped.
[0:11:42.5] David Folwell: Yeah.
[0:11:42.9] Jonathan Kestenbaum: Like, this is the thing that people miss. It’s like, we think of the way staffing works today as like good.
[0:11:51.6] David Folwell: Yeah.
[0:11:51.8] Jonathan Kestenbaum: Well, like, what if, what if actually, it actually isn’t good now? What if it actually, the amount of inefficiency that we have, I mean, like, there’s a thousand places I could point to inefficiency, I mean, just look at redeployment rates of contract labor? I mean, that’s like, I don’t even know what the percentage is today, but it’s not significantly better than I was looking at it years ago.
It was probably in the 2, 3% range back then. I can’t imagine the technology has gotten so much better there, but there’s a world where, like, every person can have a talent agent that represents them, like Ari from Entourage, you know? Like, maybe, that’s what staffing should have looked like. When you think about the high touch that that could bring and the level of service.
And the amount that that could help you as a human find your next role, you think, “Wow, maybe there is a lot more room here for technology to make this better.” And maybe, actually, this won’t just make everything disappear; we’ll be able to just provide a much better service than we did before.
[0:12:49.5] David Folwell: In theory, and we should be. I mean, I feel like automation was like a step forward, right? Like, it kind of gotten like the birthday notifications, the check-ins, and some of those basic stuff. But when you look at what AI is going to be able to do, it’s really wild. Are there specific areas, processes, or roles within staffing which ones do you think are going to change the most?
[0:13:10.1] Jonathan Kestenbaum: Listen, I live like two years in the future, usually.
[0:13:13.2] David Folwell: Yeah.
[0:13:14.3] Jonathan Kestenbaum: And so, like, it’s hard for me sometimes to come back to planet earth, and you know, generally with this –
[0:13:19.9] David Folwell: I want to go into the future, I think [crosstalk 0:13:22.6]
[0:13:23.8] Jonathan Kestenbaum: Yeah. So, I’m looking now, like, I think some of the coolest stuff I’m seeing is with digital twins.
[0:13:29.8] David Folwell: Yes.
[0:13:29.6] Jonathan Kestenbaum: It is like, essentially, like, a step beyond an agent, right? Which is where, like, I can actually train an agent in my style, with an understanding of my email addresses, my personality, the characters that I engage with on a daily basis, and it can act on my behalf as me. There’s so many fascinating derivatives that come from that. I mean, everything from, think about a situation where the obvious is, you know, it can communicate with people on behalf of me, when I’m sleeping.
But I think about other derivatives like I’m onboarding a new employee who wants to understand how my team works or what they can leverage. Well, they can communicate with my digital twin, who can share with them the information they need to know, or I’m evolving my role into doing something different, and I need someone to take over something that I was historically doing well, the same thing.
They can communicate with my digital twin, and this is stuff that I would have never had time to help someone with. I mean, building onboarding plans for employees is like the bane of my existence. So, I think that there’s a lot of really interesting new things that we’ll be able to do as a result of some of these. But you also –it’s a loaded question with a number of different pieces, you had that, you know, what –
I think, sourcing, the way we source talent, how we source talent is the first thing to change. I mean, admin. I mean, if you think about it, I don’t think we necessarily will need people scheduling interviews necessarily in the future. I mean, that’s all going to be done leveraging technology, and I think most administrative tasks within sourcing and recruiting will be done by machines.
[0:15:06.5] David Folwell: Yeah, I get excited at the thought of that, and on the digital twin side, people think that’s such a crazy concept, but I’m like, we are, GenAI and OpenAI are probably building a tool that’s going to – you’re going to be able to send it to meetings, it’s going to talk to people. Like, it’s going [crosstalk 0:15:21.3]
[0:15:22.8] Jonathan Kestenbaum: I think GPT 5 is due to come out like any day now, as reported.
[0:15:25.7] David Folwell: Yeah.
[0:15:27.5] Jonathan Kestenbaum: Yeah, I mean, like, I might regret suggesting that because you know, like, you don’t know what you read online anymore, but I – from what I hear it is, but I mean, some of these voice-spaced AI agents, I mean, that’s insane. I could take your voice and you know, have a conversation with someone and they wouldn’t know it’s you, and I think what that opens up down the line is issues around identity.
And these are like the things that I think, it’s like, you poke here, and then this side, you know, comes out, like that’s what’s kind of going to start to happen. We’re going to have now a whole identity problem to solve, and I predicted this early in ‘24, which is like, blockchain makes a comeback around identity because I’m not particularly excited about the cryptocurrency wave with blockchain.
But I think as a solution for identity, and knowing that, “Hey, this is actually David, who has interviewed for this role or who I’m communicating with,” I think that’s where that’s going to come back.
[0:16:19.0] David Folwell: It is crazy. The ability to replicate. I don’t know if replicating my voice the other day, just playing around, I was looking at Elevenlabs, and I did –
[0:16:25.7] Jonathan Kestenbaum: No, it’s fantastic.
[0:16:26.9] David Folwell: I sent a text to my mom, I went to voice and she’s like, I – she didn’t know it wasn’t me.
[0:16:33.8] Jonathan Kestenbaum: That’s awesome.
[0:16:34.7] David Folwell: So, let’s jump into kind of the next section here. You’ve touched on this a little bit, but there is the legal landmines that kind of just, you know, compliance. What’s your perspective? Any, probably a unique one as an attorney, like, what are some of the risks that you see that could come about with the future use of AI?
[0:16:51.9] Jonathan Kestenbaum: Yeah. This is like, you know, it’s my worst fear, like, the legal side of me and my talent acquisition knowledge, all coming together into this perfect storm of you’re the perfect fit to enter these legal questions. I thought I’d never do legal work again.
So, the law is basically trying to catch up with the technology, and ultimately, what the government is saying like, “Look, this is not your get out of jail free card. This is not your yacht in the Cayman Islands that you can cheat on your taxes with.”
These AI agents need to comply with federal civil rights laws or state laws, and I’m saying that in the context of US legislation, but the same would apply, you know, outside of the US, right? And so, they’re not necessarily creating new rules that AI has to follow that humans don’t; they’re creating a framework that you have to follow to make sure that you’re not violating existing laws.
And that’s essentially what the framework that they created is meant to do is to force compliance with laws. Now, I think, what people always lean to and think about is the bias side of things, which is, is this technology biased or not? Now, my opinion, that’s like 10% of the challenge. You know, that’s a binary problem that can be solved using a mathematical model, like, is bias being perpetuated in this process? Fun.
The harder part is training your entire staff on how to use these technologies in practice. So, I’ll give you an example. If I was to buy a social search tool that you know, these aggregators that go out and scrape publicly available information, and build profiles of candidates, and I was to create a diversity filter and use that information to hire an employee, I’m violating federal civil rights laws when I do that.
So, the harder part of this is how do we now train all of our people how to compliantly use AI and then build the checks and balances in place around these systems to do that, and so that’s where I think companies are going to stall in their adoption of AI. Aside from just the selection issue that I mentioned earlier, and how difficult it’s going to be to adopt it, scale, and actually get efficiencies out of it, it’s this other challenge of how do I use it compliantly with laws.
[0:19:11.7] David Folwell: Absolutely, and in practice, are you seeing any staffing firms, any corporations you feel like have a good playbook for that, or you think it’s still being built?
[0:19:21.8] Jonathan Kestenbaum: Listen, at AMS, we’re doing obviously a lot to make sure not only that we have the framework to be able to apply AI meaningfully, but also building in the guardrails to protect the AI that we do have deployed, and then what we’re really trying to do is lead the market in AI compliance. Like, we have created, you know, an advisory board of folks that are outside the industry that are helping us understand what best-in-class looks like, and hopefully creating some whatever framework that other companies can follow to deploy best-in-class AI in talent acquisitions. But, you know, we’re not the only ones doing great work there. There is a lot of folks, and I would argue also the vendors themselves are trying to stay ahead of this. There is a lot of pretty progressive vendors out there as well.
[0:20:04.5] David Folwell: Yeah, that’s great, and we’re going to jump back over on the use case side just a little bit. So, I know you are thinking two years ahead of me and probably operating two years ahead as well. When it comes to your personal or, I guess, professional or personally, what are some of the top use cases we are seeing the biggest leverage for your day-to-day?
[0:20:24.9] Jonathan Kestenbaum: Yes. So we’re, you know, we’re looking at voice-based AI agents. Typically, those are really effective in high-volume or early-career kind of situations.
[0:20:34.1] David Folwell: Yeah.
[0:20:34.4] Jonathan Kestenbaum: And sometimes it’s not just voice-based agents, it’s text-based in your communication. We are looking at building more sophisticated agentic workflows, trying to think of a good example to give you. So, one of the challenges that you’re going to find is you have 8 to 12-point solutions that support your processes in staffing, right?
Well, each of those point solutions that sits on their own unique dataset, their own part of the process is going to launch their own AI agents, and if you thought it was bad now, it’s about to get worse, like candidate experience, hiring manager experience, recruiter experience because everyone is going to be like in this turf work where whose agent’s going to win, and no one is going to have all of the data sitting in one place.
And so, one of the things that AMS is working on is building what we’re calling orchestration technology, which basically allows us to centralize the data from the disparate systems into a data lake, and then actually run AI agents across that to orchestrate tasks as to, “This task should be done by human, this task should be done by an AI agent,” and you know, start to understand and optimize the process across disparate systems.
And make the data kind of flow seamlessly, so where we’re investing a lot of time right now, is helping organizations optimize technology that they have.
[0:21:56.8] David Folwell: Yeah, and then also that’s really smart. I mean, think about having the standardization of the data, somebody who wants to use AI, want to use automation, and then if their data is not good, if you don’t have clarity around it, or if you have 14 different tools messaging from it –
[0:22:11.4] Jonathan Kestenbaum: 100%.
[0:22:12.4] David Folwell: Yeah.
[0:22:12.9] Jonathan Kestenbaum: It’s only as good as the data.
[0:22:14.2] David Folwell: Yeah, yeah, and so you guys you’re helping with customers, actually kind of standardizing that, getting into a data lake, and then actually the strategic operation on top of it as well.
[0:22:23.9] Jonathan Kestenbaum: Yes, applying AI across that, 100%.
[0:22:27.2] David Folwell: That’s great, that’s great. So, jumping back over to the future of work, you compare the future employment to WeWork in real estate. Can you paint the picture around that a little bit? This was from a previous conversation you and I had.
[0:22:39.8] Jonathan Kestenbaum: Yeah, yeah-yeah-yeah. So, there’s been stats about contract staffing growing, you know, RPO growing like the caterer of the industry, right? The way that I let that resonate with me and bring it down to like, planet earth, rather than just a piece of information on a spreadsheet, is this idea that we were able to take real estate off our balance sheet, and outsource to WeWork on a monthly basis.
And you know, that really helped in helping us scale up and scale down without having to carry this massive burden of a piece of real estate that we may or may not need over time, and I think the same thing applies to humans. Like, I think the company of the future, we don’t necessarily need full-time employees, and I think ultimately, that’s going to become pretty obvious when you have a centralized AI brain that understands what everyone does.
And who talks to who, and if you could imagine it like plugging into the machine of that company, I should be able to access pretty easily information on who I should get information from, what information I need to be super able to do the job that I was given, and ultimately, I think we could take human capital off our balance sheet, and rent it on a basis that we need it. So, I think the company of the future is going to be much simpler.
You want to own your customer relationships, you want to own your intellectual property, you want to, you know, there will be certain things that are strategic to different companies in different organizations. Like, maybe you do want to own your manufacturing facility because it’s strategic to the nature of the work that you do, but I think the large majority of service-based businesses don’t need to own a lot of things, and you know, they’ll be much simpler, much smaller, leaner, meaner organizations.
[0:24:21.3] David Folwell: So, leaner organizations, what does it mean from a staffing perspective, and it sounds like you’re basically producing at the upside if operated correctly, but how does that play out?
[0:24:32.7] Jonathan Kestenbaum: Yeah, I think you’re going to rely on contract staffing a lot more or outsourcing recruitment a lot more. Like to me, that is the future of how you get human capital engaged in organizations, and then the question which we started talking about earlier, which is like, what is the future of this look like, what could good look like? It’s like, well, every person on planet earth could have like an Ari Gold sitting in their pocket, and helping them find jobs.
[0:24:54.9] David Folwell: Yeah, that actually cares.
[0:24:57.0] Jonathan Kestenbaum: That actually cares, that represents them, that doesn’t necessarily have to make 15% even though the manager makes that, you know, whatever percentage of their salary that they make. I think that there’s really a future where something like that could happen.
[0:25:09.1] David Folwell: And you have mentioned on our previous calls what, like – I think the one thing I’ve been talking about with peers is just the evolving value chain for staffing firms, and what that’s going to look like in the future, and what the customers are actually going to be buying from the staffing firms. Like, what, you know, there is the employment or record element, but also you even brought some elements up about, you know, relationships, community. Can you state a little bit deeper into what you think will be of value?
[0:25:39.1] Jonathan Kestenbaum: Yeah. Well, like EORs, is just another good example of take this headache off my balance sheet.
[0:25:44.1] David Folwell: Yep.
[0:25:44.4] Jonathan Kestenbaum: You deal with it, you deal with the risk, and that works, and we’re doing – seeing a lot of that. I think that, as a result of some of the more administrative tasks within recruitment being done by AI, we’re going to start to move into higher-value work as recruiters to support our clients. So, does that mean workforce planning and helping them identify who they should be hiring, where they should be hiring from?
Like, I mean, imagine a world where I’m the hiring manager, I put up a role, and then you know, my staffing firm says, “Well, if you change this word in the JD, we can get you the hire in a week instead of six weeks, and actually, it will be 30,000 dollars cheaper if you hire a contractor in this location.” And you’re just as qualified for the role. So, like you know, could we start to? I think we’d be forced to.
But do more strategic work in the value chain, at the same time, the noise between staffing firms and candidates is going to get louder, as you know, we continue to. More and more staffing, it’s like the barrier is going to continue to go down, which is going to mean that the burden for us to build better and tighter relationships with candidates is going to go up.
[0:26:56.1] David Folwell: Yeah.
[0:26:57.1] Jonathan Kestenbaum: So, I believe that we’re going to have to spend more time building relationships with candidates, adding value that goes beyond just potentially what job we can get them. I think that there’s a lot that will change as a result of AI, but I think the challenges, I’m definitely not going to get it 100% right, and so you got to take small baby steps to kind of evolve as things evolve around you.
[0:27:20.1] David Folwell: Yeah, and it’s always so that it’s shifting to your – towards strategic partnerships, which I think is going to be a theme as AI kind of commoditizes the labor side of that. I just think about what HR used to buy. One of the value components of a staffing firm was that you had somebody who’s willing to do a hundred calls a day, you know? That was it, that was like a key part of –
[0:27:41.8] Jonathan Kestenbaum: Yeah. I don’t think you have that anymore.
[0:27:43.7] David Folwell: That and now –
[0:27:45.1] Jonathan Kestenbaum: And look, I want to throw another dynamic I think is super interesting. Maybe we’re not only in charge of hiring human capital, maybe like staffing firms are in charge of hiring these agents, because like, if you think about it right now, companies, individual business units within companies, are buying software that does the work or tasks of humans in some capacity when they’re buying AI.
There’s going to be a significant amount of overlap between two technologies that two different business units buy. This is no one centralizing it; TA is not involved in that decision-making process, so therefore, you have technology that could be violating labor laws, or decisions have to be made around people as a result of this technology that I started talking about, skills-based firing, you know? Because this skill is no longer even needed in the organization, you know?
So, like I think that someone is going to have to centralize that buying of AI in a way that you can kind of see impact it’s going to have on different human capital in the organization, and then also make sure that you are not paying twice for the same functionality.
[0:28:51.7] David Folwell: That is funny to think about in AI, with a staffing agency that procures and manages your agents.
[0:28:59.0] Jonathan Kestenbaum: Yeah, and also understands who does what well, you know, which agents can be effective where.
[0:29:04.6] David Folwell: Yeah, we’re going to need something like that; that role is going to have to exist as the proliferation agents come to that. I think that it’s right now, we’re just at the early stages, and we are already seeing the overlap, which is crazy. So, what are some of the competitive notes if you’re a staffing agency? What would you be doubling down on right now?
[0:29:21.9] Jonathan Kestenbaum: You asked me the hard question that I haven’t had time to operational as I’m thinking around, but I’ll share my thoughts, I guess, anecdotally. Like, I think you have to start to think about how you leverage AI. This is not a game where you can sit on the sidelines and wait. You have to be in the game because your competitors will get significant efficiencies as a result of AI.
The second thing is the process of having to sell to both candidates and organizations is not going away. That’s the last thing that’s going to get automated. So, continuing to build differentiated value for your clients, I think being able to accomplish business problems for them faster, either candidate in the seat quicker, or a better quality candidate in the seat, I think, you know, these things don’t go away as a result of AI. So, continuing to focus on sales and building deeper relationships with candidates is key.
[0:30:16.5] David Folwell: Yeah, I couldn’t agree more, and from a tech perspective, I know a lot of the agencies are trying to figure out where, you know, with budgets, and what to do, I hear constantly that there’s just too many options from a tech perspective, that there’s too many vendors out there, too many AI choices, and I think people are kind of confused. Any guidance on, and obviously, I told them to go look at the talent tech ecosystem as the way to kind of navigate that.
But any guidance in terms of how they should be thinking about our approach in their tech stack strategy?
[0:30:49.8] Jonathan Kestenbaum: Yeah, I mean, call me. Every company is different, right? It’s really interesting. Like, I was inside one of the world’s largest bank actually last week and we were having a conversation about the head of talent opinion was, “If we buy AI, and the candidates are using AI, this is just AI talking to AI, and then, I have the same problem as I have now, but just at a larger scale, you know?”
So, his kind of “aha” was, “Well, maybe I should be focusing on how I assess candidates more meaningfully, and filled AI and unique processes around how we assess candidates, than trying to figure out how we use AI to source more candidates because the candidates have access to this technology as well.” So, I thought it was just an interesting kind of reframe for me, but yeah, I think that you have to understand this dynamic of it’s not as simple as just selecting the right technology anymore.
It’s about selecting every technology, doing the change in management around the technology, and then really looking at the task that the humans you have around the technology you’re doing, and continuously, on an ongoing basis, transforming those people, and I’ve been saying, “Instead of BAU, it’s TAU, transformation as usual.” So that you can optimize the process around the technologies as they evolve.
And that is, becoming an organization that operates in TAU, instead of BAU is that totally new muscle that has to be built.
[0:32:12.9] David Folwell: Yeah, it makes a ton of sense. I feel like, on the adoption of this also, I think that there is getting it down to the ground level, have you seen anything? You know, I’ve heard one organization where there were, “So, we’re requiring everybody in our team to just use AI twice a day.” Other than that, the core refund.
[0:32:29.0] Jonathan Kestenbaum: Yeah. You know what I do with my team? I ask them on a weekly basis, “How did you use AI this week, and what did you use it for?” And they have to share it, right? With the team, and there’s always someone who has a unique use case, and you learn from them.
[0:32:43.6] David Folwell: Yeah. That’s great, and I think that’s it, and also puts the people who know that questions are going to come up, and they know they’re going to have to go around and answer, which means that they’ve got to use it, which is great.
[0:32:53.4] Jonathan Kestenbaum: Yeah, because like, in my opinion, like, people, it’s like using Ozempic, people feel like it’s cheating, right? Like, I don’t think it’s cheating, it’s all about leveraging it to make yourself better, right? So, like, as long as you’re not just using it and not reading the content, not optimizing how you frame your prompt, I see no issue with it.
[0:33:13.7] David Folwell: When people consider it cheating, I always think back to when Google first came out, and I was like, “Should I not Google it? Should I go to the library and like, do I need to know the Dewey Decimal System?”
[0:33:24.4] Jonathan Kestenbaum: Yeah.
[0:33:24.9] David Folwell: Anymore? Like, it’s a tool, it exists, other people are using it, and it can make you more efficient, more productive, in a pretty significant way. So, just a couple of last questions for you. So, I want to know if there’s anything about AI that you just think staffing firms get wrong or that you see businesses get wrong? Any misconceptions, or ideas around it that are – seem off compared to what you – how you see the future? As a chief, you know, would be a good example of what one, so.
[0:33:55.4] Jonathan Kestenbaum: No, I think it’s very easy to be wowed by some buzzy cool feature that some technology has, but if you can operationalize this into your business process, it’s not going to work.
[0:34:06.8] David Folwell: Yeah.
[0:34:06.8] Jonathan Kestenbaum: And so, it’s super important to realize that the technology is only, still, only as good as the people who operate it, and the amount of work you put into operationalizing it into a process that’s seamless, because you need your people to climb a mountain to get access to this tool to leverage it, like, it ain’t happening.
[0:34:26.3] David Folwell: Yeah. The behavior change is hard.
[0:34:29.7] Jonathan Kestenbaum: Yeah, a hundred percent. So, I think, you have to really work twice as hard at identifying the business problem first and making sure that you optimize the process around the technology or the process with the technology, and the selection is just a small piece of that.
[0:34:48.1] David Folwell: Absolutely, absolutely. I got two last questions for you, and then we’ll close out some final comments, and this is kind of what we call the speed round here. So, the first question I got for you is what book or books have you given most as a gift, and why?
[0:35:01.3] Jonathan Kestenbaum: What book or books have I given most as a gift?
[0:35:04.2] David Folwell: Or have been most influential on your life?
[0:35:07.3] Jonathan Kestenbaum: I love, I absolutely love to read biographies of entrepreneurs. This is like, some people study athletes, I was never the guy that could tell you the stats of any player. I mean, I did, actually, by the way, side note, write Jeremy Lin’s biography and published it in 2012, but that’s not because I’m an athlete or was particularly interested in sports, it was more of an entrepreneurial endeavor.
So, I love reading biographies. I think, probably, Walt Disney’s biography is the one that’s had the biggest impact on me, and you know, something that I think everyone should read. But I’ve equally read Jeff Bezos, Elon Musk, Thomas Edison. I mean, these are just people that I look up to.
[0:35:48.7] David Folwell: Yup. That’s amazing, and Walt Disney one is one I need to pick up. I’ve heard multiple people say it’s incredible.
[0:35:55.0] Jonathan Kestenbaum: Incredible, yeah.
[0:35:56.1] David Folwell: Yeah, yeah, and what habits have most changed your life in the last one to three years?
[0:36:03.9] Jonathan Kestenbaum: I work out every single day, that for sure has changed my life. Waking up in the morning, 5:30, going to the gym, getting my heart moving, being plugged in before anyone else is emailing me, and catching up on my emails, like that, for sure has changed my life.
[0:36:18.3] David Folwell: Love it, and this one might be kind of follow suit, but what is the best or most worthwhile investment you’ve ever made?
[0:36:25.0] Jonathan Kestenbaum: Oh, my wife.
[0:36:26.5] David Folwell: Yeah, your wife?
[0:36:27.5] Jonathan Kestenbaum: Having my wife. Oh yeah, absolutely, for sure.
[0:36:30.9] David Folwell: Love it, and with that, I wanted to just ask if you have any closing comments to the audience?
[0:36:36.5] Jonathan Kestenbaum: I’ll just joke, to go back to this investment one, it’s definitely my wife, but like, I’m the king of buying electronic things, you know, like any new electronic, I buy, and it’s always, it’s like, one out of 10 of these new electronics ever actually make a meaningful impact or change in my life, and you were – we’re talking like, the Apple watch, you know –
[0:36:53.7] David Folwell: Yeah.
[0:36:53.7] Jonathan Kestenbaum: Aura ring, whatever the new tech thing is, like I’d buy it, right? But as for that one out of 10 times that it changes my life, this watch that I have, this addicted thing where I buy the next thing, you know? But yeah, none of those physical things actually –
[0:37:08.7] David Folwell: You identify what’s next. I’m right there with you. Number of paid subscriptions that I have at the end of every month, and I’m like –
[0:37:16.4] Jonathan Kestenbaum: Embarrassing.
[0:37:17.7] David Folwell: Have to try it.
[0:37:20.1] Jonathan Kestenbaum: Totally embarrassing.
[0:37:22.1] David Folwell: Any closing comments for the audience?
[0:37:24.1] Jonathan Kestenbaum: No, listen, I think this is the most exciting time to be in recruitment. I think that the opportunity is going to grow. I think that, obviously, change is scary, but we have an opportunity to make the things that we do have a bigger impact, start to become a more strategic part of an organization, and you know, I think it’s exciting.
[0:37:42.1] David Folwell: I absolutely agree with that, and Jonathan, I really enjoyed having you on the show. Some of the deeper, more forward-looking thoughts we’ve had around AI. Love the idea of the digital twin, I think it’s coming, and it was great having you on. I hope you have a great day.
[0:37:56.3] Jonathan Kestenbaum: You too, thank you.
