Why Does AI Momentum Stall in Large Organizations? with Kyle Lagunas & Allyn Bailey

(Intro Narrator at 00:00:00) Today, we're talking to Kyle Lagunas, analyst and founder at Kyle and Co, and Allyn Bailey, senior director of communications at SmartRecruiters, about why AI momentum is stalling in organizations. You're listening to Joel Beasley, Modern CTO.

(Joel Beasley at 00:00:21) So Kyle, you did some research. The AI momentum model—was that what you were hoping to get out of it, or is that something that came out of it after you already started?

(Kyle Lagunas at 00:00:33) Out of it, actually. Yeah. I'm sure you've seen a number of maturity models through your career and through this pod, and that was originally the thought process. But we started getting in and we recognized AI doesn't exist, or maturity doesn't exist in AI yet. Why are we chasing something like that?

(Kyle Lagunas at 00:00:54) And instead, what we observed was the people who continued to maintain momentum in AI innovation were the ones that were out in front, which in a sense, they were seeing more results, getting better buy-in, better alignment. Maybe now we'll have a maturity model, but at the moment, I just wanted to see how people kept moving.

(Joel Beasley at 00:01:16) We're still on the momentum model.

(Kyle Lagunas at 00:01:18) I think so. Yeah, I think so.

(Joel Beasley at 00:01:21) And then Allyn, you're with SmartRecruiters, correct? So you guys got together—SmartRecruiters and Kyle and Co got together—and you surveyed 350 HR leaders and then built this AI momentum framework. Is that right?

(Allyn Bailey at 00:01:35) Yeah, absolutely. So really, our role in this was to say, listen, we know as we work with customers and prospects and people who are in the talent acquisition space and the larger HR space, they're constantly coming to us trying to understand what they need to be doing with AI. What works? What doesn't work? What is ethical? What's not ethical? What's appropriate? What's not appropriate?

(Allyn Bailey at 00:01:56) There's a lot of questions, right? And not a lot of answers. And we realized that one of the best things we could do for our customers was start to get a sense for, are there some commonalities? Are there some lessons that can be learned? Are there some things that we can bring to them that helps them be able to put their arms around this and figure out how to move a path forward? So that was really the intent. And I like where Kyle went, framing up this idea that we did. We walked in with the assumption saying there's an in-state, there's a maturity state.

(Allyn Bailey at 00:02:32) And I think realizing very quickly that there wasn't, and that if it's about momentum, then how do you build momentum, and what are the criteria for that? And Kyle's team and Kyle did a fabulous job really starting to outline those key components that actually create that momentum within the market and for an individual company.

(Joel Beasley at 00:02:54) That's interesting too, because you guys are in the HR space, which is unique to people. And the ethics gap is actually a great conversation. So you looked at it more as what are the different companies doing and how they're handling this, because there's not really an independent ethics company that's niche for HR-specific things. Right? I mean, I'm assuming that there are massive AI ethics councils out there that are broadly saying stuff, but it's so specific to the exact use case and scenarios. So you're just taking the approach to the org and the organization?

(Kyle Lagunas at 00:03:26) Yeah, and to the organization and its operating culture and its risk management profile and its governance. HR is, oh my God, it's a really exciting space for AI because there are just so many really neat use cases. But I think it's actually been really cool to start looking at what is helping companies move forward and move faster, or even just continue to move with AI innovation. And it's actually the most boring red tape stuff that is helping them, and not just big ideas and CFOs willing to cut a check. I think that was what was neat about this too—we have more questions than answers. Joel, you talk about ethics, and it is such a subjective thing. We don't even really have a ton of regulatory compliance for it specifically for AI's use in employment either.

(Kyle Lagunas at 00:04:35) So people have a lot of questions. What was neat about this was to find, hey, y'all, the stuff that's helping is stuff that has always been important for modern, tech-enabled enterprise. And that's almost like a relief of, oh, well, so these nine drivers—we can manage these. I can work on these. I can iterate on these. I don't have to just run ahead blindly and just go all in on AI. There's stuff that I can do, stuff that I can get alignment on, and things I can chip away at.

(Allyn Bailey at 00:05:13) Yeah, I think that's the critical piece, right? The ability to have a clear path forward that says rather than feeling a sense of anxiety, being stuck in place, how can I take—how can I, as a leader in the HR space, a leader in the TA space, or even a leader in technology in general for my organization—how can I take some autonomy over the decision making? How can I start to build towards something that I can feel comfortable with and not be afraid when I give the answer that everybody else is asking me for? Right? Because HR fundamentally was built and framed as a risk management organization. Right? So at its core and at its root is this idea that if we bring something forward, the job is basically, in not-so-fancy words, don't let the company get sued.

(Allyn Bailey at 00:06:07) Right? So I think that in conjunction with this world of the unknown of AI was a real big opportunity space to help give them some tools.

(Joel Beasley at 00:06:18) And so how can an organization that has AI ambition use this report to actually gain momentum?

(Kyle Lagunas at 00:06:25) It starts with know thyself, right? I love that proverb, know thyself. That's where so much of this begins. And I think it is understanding where you are. And people immediately go to how many different AI features have we turned on and how many AI pilots are we running, and it's not that—and it could be. But what we ended up doing was creating an assessment tool. Because we went through the survey, and I don't know if you've ever participated in a primary research project, but you start with asking way more questions than you need, than you actually land on. Right? Like the answers to. So we had, I don't know, 40 questions in our survey, and we found that there were really just 12 that we really needed answers to to get an understanding of where you're at. And so we built an assessment tool, nothing fancy—with Typeform and some email automation on it, nothing crazy.

(Kyle Lagunas at 00:07:40) But where you tell us about, where does, who owns AI governance in your organization? Is there HR-specific AI governance? Do you have dedicated budget for AI projects in HR? The one that was really interesting to me was, how literate is your entire HR organization when it comes to AI? Not your special projects people. And so just by answering a few, I think, relatively straightforward questions, we could give you a sense of your momentum archetype. The other thing is, although momentum implies a movement in one direction, momentum is multidimensional, and it's not an up-and-to-the-right, maturity thing. So we were able to also give people a sense of, here's where you're really strong, here's where you are not so strong, and here are some places that we think you could start shoring things up.

(Kyle Lagunas at 00:08:31) Because HR can't just overnight get a bunch of budget. You know, investment is one of the drivers. We do need some investment. And they can't completely change an organization's posture when it comes to risk. But they can begin to build literacy without pilots just by accessing all of the content on the web. So yeah, it really is a really simple thing. You wouldn't know it because I think our report was 60 pages long, but that's what happens when you give a researcher free rein. But the tool itself is really lightweight, and it's designed to be a starting point because I don't do consulting on this. As a—that's the funnest part about being a researcher—I just get to find answers to the complicated problems and then let people do with that information what they will. But yeah, I think that the assessment tool has just been really helpful for folks. We run workshops with it with HR execs and get insane feedback on it as just a really good starting point for them to bring back to their teams.

(Allyn Bailey at 00:09:36) I think the other thing that that tool does is it helps create some commonality in language. So now as a CHRO, as a TA leader in this space, I can bring forward some language and some frameworks that help people align to as I'm talking to my CTOs, I'm talking to my CIOs, I'm talking to my CFO. And I can frame for them not just what's happening, but some language we can all align around. We start talking about commonalities and language and tone, and then commonalities around what goodness starts to look like, right? So we can all benchmark there. Because I think sometimes in organizations that we work with, there's a preconceived notion that, you know, yes, I may be the CHRO of an organization, I'm gonna go in and assume that the CTO or the CIO has some grander picture understanding than perhaps I do of this space. And one of the things we're really working with our customers and our prospects on is helping them understand that they're bringing a very unique perspective and understanding of the larger HR picture, landscape, framework, guidance, goodness, and best practices.

(Allyn Bailey at 00:10:54) And without that lens, when they go in and have that conversation with the rest of the C-suite, it becomes really easy to lose track of the human center piece of this whole AI navigation that we're going through. So giving them some common language there, helping them understand where they need to bring in their unique perspective into the dialogue, and where they need to be leaning in and supporting this larger framework.

(Joel Beasley at 00:11:20) Now going into it, was that what you were hoping—when you were looking at SmartRecruiters making this decision to back this research, are those the reasons why, or did you find them out afterwards?

(Allyn Bailey at 00:11:31) Oh, I think there were some pieces of this. I think the biggest—we were really trying to answer a question just to start out with. And the question was, with all of the noise out there and all of the demand for AI-based products and "show me what your AI technology can do," we were still seeing organizations stagnate, not actually moving forward. So there was a lot of asking, not a lot of actually putting things into place, implementing, managing. And so we really started this with the foundation that says we have an open question, which says we're a business.

(Allyn Bailey at 00:12:10) We build these technologies and these tools. Besides believing they're the right direction for many companies to move in, fundamentally, we want to be able to understand, how do we get over that hurdle of angst? Or is it angst? Is it a change management problem? Is it a technology integration problem? What is stopping that momentum and that movement forward? And so that was where we started. That was the root. And once we started to identify that, then we branched off from there and said, okay, if this is the root, which is basically a lack of commonality of language, a lack of understanding of what to do next—you know, sometimes feeling stuck isn't just "I don't want to," it's "I don't know what to do." Once we figured that out, then we knew really quickly it was about how do we work with Kyle and team to be able to frame that into, here's what you can do.

(Joel Beasley at 00:13:01) And then Kyle, the model has three dimensions—capability, posture, and investment. Can you explain those to me?

(Kyle Lagunas at 00:13:07) The three different dimensions of the model are, as you noted, capability, posture, and investment. Investment is super clear, right? It's, do we have resources? And not just dollar resources or euro or whatever, but also headcount resources. Are we dedicating specialized headcount to projects? Are we investing resources in making sure that the rest of the enterprise operation is going to support these projects? Are we aligned on these things? So investment's easy. Posture is almost like point of view.

(Kyle Lagunas at 00:13:40) So, are we—do we remain committed to expanding the use of AI in HR? And so that's one. In quantifying that, I looked year over year, just their willingness to continue to add in new pilots and programs. The other is risk, which is really important. What is your appetite for risk? How do you think about risk? For HR, it is a really special one because risk—our favorite method of risk management is risk avoidance. And what happens when you are just so used to risk avoidance is you never really learn how to manage and mitigate risk and how to actually calculate it. You just say, there's risk here, I'm not gonna do it.

(Kyle Lagunas at 00:14:30) Well, guess what? That really doesn't work for AI. And then the other one was governance, because I think that governance and risk go very much hand in hand. But governance is the rules of the road, right? That's the guardrails that you put in place. And yeah, the capability part is stuff that you would probably expect—integration of systems and data and processes. The one that I put in there that we are continuing to track, and it's probably one of the things that keeps me in business, literacy—AI literacy—is one of the biggest make-or-break capabilities for those HR organizations that have been able to maintain momentum and those that are falling apart after one pilot doesn't go the way that they had hoped it would. So yeah, we tried not to—I'm trying not to come up with something cute with this model.

(Kyle Lagunas at 00:15:32) I was trying to put something together that really did reflect the current experiences and realities of HR and talent leaders. Because I mean, they don't want just another random model. So it was kind of hard to write thought leadership about things like governance and risk. But then as we brought this research out into the market and we did a workshop with the SuccessFactors SmartRecruiters CAB in the US and in the UK—and the sampling for the survey was two-thirds North America actually, and two-thirds UK, EU—and it was really neat to have conversations with them. Most of the conversation ended up talking about how they can manage and navigate risk and governance when they don't control those programs, when they're stakeholders for those enterprise programs.

(Kyle Lagunas at 00:16:24) And they were finding that was what they were trying to angle for more at the AI councils that they get to contribute to or be a part of. So yeah, putting this together was like, is this gonna be cool enough for people to read? Because I'm interested in this, but it's getting kinda dry. And then sure enough, as soon as we sit down with leaders, they're like, give us more of this. Go into more detail here. This is the part that is a major blind spot for me. I know I need AI literacy. I know that I need to continue to maintain investment in AI programs. It's these other things that are really difficult for me to influence because I don't own.

(Joel Beasley at 00:17:04) Yeah. The literacy is a big thing because if you don't have the words to even have the conversation, then how are you gonna make any change at all?

(Kyle Lagunas at 00:17:13) That's exactly right. How can you even be a part of the conversation if you can't speak the language? And that's something that I observe as a big risk factor for HR in that we had typically leaned deeper into HR domain expertise and away from technology and data expertise. And that is leaving us vulnerable in this wave of transformation because we are starting from much further behind than some of our colleagues are. And I think that, or I even observe that we have challenges building inroads with our partners and co-innovators, like the CTO's office, because we have gone so far with that gap between our literacy level and theirs. And it's, I think, the biggest hurdle, the first hurdle that we really need to clear. Allyn, I know that you and I worked together. Joel, I've been on the practitioner side and on the vendor side, just like Allyn. We've just loved this ecosystem, so we keep fluttering around it. But during COVID, as a story about literacy, during COVID, we had blank checks to buy tech. It was like, if you had a problem, you could throw a body at it. You could throw a tool at it. You could do whatever you wanted. And coming out of that, and we did, we bought a lot, we hired a lot. When we came out of COVID, there, it was really scary because there was just so many layoffs and so many cuts everywhere, but you kind of forgot for a moment. We got bloated, extremely bloated with these tools and with these programs and all these really great ideas. The business said, sure, do it. We'll fund it. We'll fund it. We'll fund it. Well, because we didn't actually have the capability in the HR and talent organization to effectively manage those programs and effectively drive change management around new ways of working. It was just more stuff that we had in our stack and no real improvements in the outcomes of our functions. And so now, that was just a couple of years ago. The cycles are moving so fast. Now, things are moving again and we're starting from this point where not only do we not have the data and tech literacy, we can barely spell LLM, but we also have not the best reputation for really doing a great job with the tech innovation that we have gotten the permission for in the past. I think that's partly why I was so excited to partner with Allyn's team on this is because we care. She and I really care about these people. We really want them to be heroes. We're trying to find meaningful answers to help them overcome this.

(Joel Beasley at 00:20:01) When I was reading, I found that and I was reading the report. It said 34% of the HR orgs are just exploring AI. Only 12% have strategically accelerated. Have you looked deeper at those 12% to see what they actually did, what areas they accelerated?

(Kyle Lagunas at 00:20:17) So I didn't, because I started looking at the qualities of those organizations. Because as I looked at the use cases, I asked everybody what use cases have you tried in the last twelve months and which ones are you piloting in the next. And candidly, I didn't really see a lot of, oh, these are the on-ramps. These are the pilots and use cases that our people are starting with. It really just depends on the business problems. That was actually how I got to these drivers in momentum. I was like, well, what are they doing then? What is different about the companies that are out in front versus those that are stuck? And I'm looking at the qualities of them. I'm like, alright. Well, they have HR-specific AI governance. They have improved AI literacy. They have deeper collaboration with procurement and IT and legal. And so that's where I was like, oh, this is not gonna be very cool. But it's gonna be really useful because I'm finding it's not the cool stuff that they're doing. It's the boring stuff that is gumming everything else up that they've already solved for or have made more progress on.

(Allyn Bailey at 00:21:29) I think that's such a critical point because I think so many people that we're working with come into this believing that, to move forward, they're sold a concept or an idea. Vendor comes in, tells them you could do X, Y, and Z, or here's this cool thing. Right? So it's about the new thing. Am I behind? Do I need, oh my goodness. Do we all need to learn Claude now? We don't know what that is. Right? So they get very hyper-fixated on what they think this new thing is over here. And, fundamentally, at the end of the day, it's all about core infrastructure, understanding that piece. Right? It is, do you understand your compliance? Do you understand your data pathing? Do you have mechanisms to measure success? Do you have frameworks that allow you to go in and make shifts inside your tools, an open box or closed box. Right? These, from a CTO's perspective, they're probably listening to me say these things and going, yeah, duh. Those are the things that you need to have. Right? From an HR perspective, just like Kyle said, we were coming from a space which was like, if it's a cool thing and they're selling me on a good usage model that's gonna solve this problem I have in the HR space at some level, then I'm gonna go for it and not necessarily think about all the consequences that are related to it. These are all the reasons why HR and CTO have always had a little bit of fun sitting across conference tables. Right?

(Joel Beasley at 00:23:06) Yeah. It almost seems like some, I bet you somewhere, some CTOs have put teams on the HR team.

(Allyn Bailey at 00:23:15) Oh, 100%. Because, yeah.

(Kyle Lagunas at 00:23:15) And in fact, whenever we do, I feel like that's a big unlock.

(Joel Beasley at 00:23:23) There's that 12%.

(Kyle Lagunas at 00:23:24) Yeah. Right. Yeah.

(Allyn Bailey at 00:23:25) That's right. There's that 12%. And it is, fundamentally, where you're right, there are the lessons in that 12%. Like Kyle said, they are the groups that are building collaborative teams together, that are working across business silos, the teams that have guardrails that they're trying to put in place. There's a really interesting piece in the research too. I keep going back to it because it fascinates me, and I think it's the story here. When we first looked at it, you know, you walk in with an assumption that says, locations in the world where there are more rules are going to have less adoption and have lower rates of AI usage and have a slower on-ramp into the space. We've very quickly realized that was not what we were seeing. Right, Kyle?

(Kyle Lagunas at 00:24:14) Yep. Yeah. Because we compared the EU-UK audience, which is, you know, subject to the EU AI Act. And we assumed that because of that massive piece of legislation, regulation coming that EU-UK would be significantly behind, and they weren't. I mean, they were pretty much on par with North America. Are they ahead?

(Allyn Bailey at 00:24:43) In many places, they were accelerating and going into different use cases because they had, because they understood what, if you go and start looking at the comments and the dialogue and that, what you're starting to see is they understood where the guardrails were so they knew what lane to play in. So now they can go play versus just it's a wild, wild west and let me try anything, and that feels overwhelming. And that's also where I get shut down if I start going to my business partners to try and get them to collaborate with me.

(Joel Beasley at 00:25:11) Well, if it's the wild west, you have the, I'm experiencing this, and I own a couple different businesses. And one of them is a crypto mining business. And so over there, what that's been like the past decade, everyone's just like, please tell us what we can do. Everyone wants to run, but we're all so scared that we're gonna do something that's gonna be retroactively a nightmare for us.

(Allyn Bailey at 00:25:38) Yeah. Exactly. Exactly. Right? Which goes back to my basic foundation, which is, you know, HR is the team that doesn't wanna get anybody sued. God forbid. Right? But at the same time, wanna get in there and do it. So if you tell me what the rules are, we'll go play. We can play with rules very, very well if we understand what they are.

(Joel Beasley at 00:25:55) Oh, yeah. As an engineer, we try to break them.

(Kyle Lagunas at 00:26:02) I think it's a critical thing for HR personally, because there's a reputation, credibility capital that is, I think, we have a deficit of and that we need, not just for our own projects, but so that if we do have a seat at the AI Council table, which, by the way, is a table I think that we should all be getting at right now. But if we are there, we can't be the ones that are like, I hate to be the bearer of bad news, everyone, but we're not gonna be able to do that, you know, because yada yada yada. Instead, it's like, interesting. This is gonna be complicated because, you know, we operate in this state, this state, and in these countries as well. And let me just take that as something for me to go and figure out and come back with some go-forward solutions, some options for us. We really need to build up credibility so that we can also meet the moment that the enterprise is chasing. Not just for us to shore up our own operations and to deliver better outcomes with the resources that we have in HR and talent, but so when we are looking at widespread workforce disruption, that we can be advisors and stewards of the business at that same moment. I worry about anyone who is leaning deeper into HR subject matter expertise at this time and not leaning forward into more solution-oriented points of view. And that's just me waxing philosophical. I'm not sitting in all of these AI Council meetings and in these boardrooms. I think it's just worthy to call out that I feel like there's a mindset shift that needs to occur for HR. And I think that embracing this AI innovation more effectively is gonna help with that a lot.

(Joel Beasley at 00:28:00) I've got this question here for you guys about the data. I'm almost scared to ask it because it's got so much behind. But the HR leaders prioritizing the quality of hiring and retention, but then getting measured on speed and compliance. Is that the same at all the organizations, or is that just, how does that work?

(Kyle Lagunas at 00:28:25) I'm so glad you caught that. Because I was wondering if I should include some of these state of HR as a part of this study, but I felt like it's important context. Allyn, I would love for your take on this because I don't think that you and I have talked about it since we first looked at the data.

(Allyn Bailey at 00:28:42) So, I mean, I think, so this is, I don't wanna go into my big history lesson of HR, but I will go just down a little bit of it. Right? If you think about what HR is, what human resources is inside an organization, it really has two components to it. One is actually the real root, and the other one is the one we like to talk about because it says the word human in it. The first and the deepest one is it is a transactional business or has been primarily. I know people are like, don't say that aloud, but it's okay. It kinda has been. Right? It's about I process paychecks. I process applications, right? I process help tickets. I deal with kind of benefits-related support questions, right? These are kind of frequent things that that arm of the business does. We've primarily called that HR operations, kind of HR ops, but it is where a big chunk of our technology frameworks have been going because that's where we put our technology and our tools to automate, to create efficiencies, etcetera, to also where we spend a lot of time doing things like moving into call center strategies five or ten years ago, live in that same space. Now there's that kind of human piece, which is the HR business partner, somebody who is collaborating with the business to think about the psychology of change management, helping businesses understand how you're putting the right people in the right spots and how that's really helping a business kind of accelerate and move forward. Transparently, though, when we're talking about the technology landscape and how we're leveraging it, almost all of it and almost all this conversation has been in that HR ops space. Why is this important? Because if you think about all those things I just measured or just talked about, how you measure or how we have measured success in those spaces has been, is it efficient? How cost effective is it? And does it allow me to move the business forward at the pace I wanna move it forward? Very transactional sort of measurements. They're also the things that are easier to measure. Human elements are much harder to measure. Do people feel better in their job? So then I end up with an engagement survey. Right? But even that is really subjective in a lot of ways. Or are we looking at retention data? Are we looking at performance data? All of these elements, which are more complicated human parts of the story, they're harder to measure, they take more time, and they take more interpretation. And so naturally, businesses lean into those efficiencies and operational data problems. Therein lies part of the challenge, which then leads us to, if that's what we're measuring on and what we've been measuring ourselves on, that means we've been putting a majority of our effort towards what are the technologies that allow me to be faster, right, and reduce the amount of overhead that it requires for me to accomplish things. So it's been like an automation free-for-all.

(Kyle Lagunas at 00:31:54) Yeah. I think it's just interesting to look at the reality. And the history lesson is really helpful. Context always is. But what we're being held accountable for versus what we say is important, when those are so different in quality, it's going to, it just, I don't know, crisis of innovation identity that we are also grappling with at the same time. And it's no surprise then that so many companies are stuck in just this exploration mode where they're looking at what they might do, which pilot, which bet they might make because they're afraid to make the wrong one. But what is, I think, interesting is if you look at, because it's easy to think that those who are, or it's tempting maybe to think that those who are focused on automation versus those that are focused on orchestration, like, agentic versus machine learning. So it's like, oh, orchestration is a way bigger word, so that must be the better thing.

(Allyn Bailey at 00:32:58) It's something much more important.

(Kyle Lagunas at 00:33:00) You know? Like, oh, that's what the mature ones are doing. I'm like, well, no. The mature ones, they're on that work now because they already started with automation of processes as soon as advanced automation became more feasible for us. But it is still, I think, telling to look at not just what HR is held accountable for, but what the primary drivers of AI investment is for HR. And efficiency and speed is the number one. That was 50% said that was one of their top. But interestingly enough, the second highest answer, second most common answer, was improving decision-making quality. So, you know, there is room for both. That's what's super cool about AI innovation is these aren't binary choices.

(Kyle Lagunas at 00:33:58) You can run several pilots in tandem. Even just this project as an example, we started with one outcome in mind. And as we delved into the work that we set out to do, we realized that there was something else there. And I think that's something that we can pursue and expect as we continue to build AI momentum, if you will, which is that we're going to find that some things that we expected to be huge hurdles maybe aren't. And there are things that maybe we already have some of the answers to, and that we can do a lot more than we ever thought was possible and do it faster than we thought we could do.

(Kyle Lagunas at 00:34:47) It is—all of this is just—the pace of this change, the scale of this change, the fact that this is cultural, operational, and technological all at the same time. This is just a lot for an organization like HR to absorb and to navigate, much less to lead. And so, yeah, we did the study five months ago now, and this research hasn't—the shelf life of this research hasn't waned whatsoever. It's just something worth validating. So that at least tells us that the goal posts aren't shifting that dramatically.

(Kyle Lagunas at 00:35:28) It's just—what's the volume of output? Because I'm sure we got several new models from Claude and ChatGPT since this went live.

(Joel Beasley at 00:35:39) I think we just got one yesterday.

(Kyle Lagunas at 00:35:42) I know. I did. Yeah. Right.

(Joel Beasley at 00:35:45) Which I'm curious to know, Allyn, you talked a little bit about how people have been using it for automation, and that's fine and that's fair. But are they using it yet as a way to think through problems?

(Allyn Bailey at 00:36:01) So, the reason I'm pausing here is I will tell you—let me go back to this. I think it depends on what, on your role and where you're sitting, and how they're leveraging that technology. I like the terminology Kyle's using around decision support. Right? And it providing access to interpretation of data to help support me at the right time and the right moment for decision support.

(Allyn Bailey at 00:36:25) I think we are just starting to understand that a lot of the things that we were putting in place or that companies are putting in place that we were thinking about as automation or as process simplification, et cetera—if adjusted or if you think about them more holistically, can actually create some of these more decision support moments. So let me give you an example. Early days in the talent acquisition landscape, chatbots were the beginning. Right? Like, okay, I'm going to use a very simplistic NLP-based bot that's going to be able to answer questions for me on my career site, et cetera. Right? Then we advanced to the next stages. We started to get LLMs. We realized they could be more complex. We could actually be more robust with them, and we started to add them into more complex conversations with candidates. Right? Then we could turn around and say, wait a minute. I can also apply this same strategy in conversational interactive strategy in what I'm working with as hiring managers. Right? So now I can take the same product, move it over to a companion place with hiring managers, and then by giving it access to different types of data pools and influencing when we want it to surface information to people, both proactively and not just reactively. Now we're moving to a place where these technologies that were put into place to do very kind of simple early transactional actions can now start to be patched together to build decision support frameworks. The challenge I'm finding with our customers when we go into this space is that it's not that it's not feasible and that they couldn't use it for these mechanisms, it's that, one, their business processes may not really be well aligned to the insertion of new inputs to help you make decisions. Right? So for example, the step process is I fill out a form to say I'm going to need a new role, and then there's a body or a person whose job it was to look at that and to assess it against the budget and decide if it's going to get approved. Well, now I can provide data using my HCM, et cetera, to that person, and maybe I don't even need that person anymore. Maybe there are decision criteria that can be put into place, and they can help support the decision-making process as we move into those next things. But my business process isn't developed for that yet, so companies are challenged with problems like that. The second piece is that there's a natural—and I'm sure we've seen it not just in the HR space. I would love to get your perspective on this, Joel. But as you start to provide insight into information, right, so kind of surface information for people to support them in a decision support space, there's still a natural tendency to either be skeptical of the information that's being provided, assume it may not be as robust or as rich as you need it to be. And so there's a little bit of tension that's kind of going on in that moment. Right? Like, can I use it to support them? That was a long-winded answer to say we're getting closer. People are starting to do it in these spaces, but there are a lot of things that have to happen. The data has to get connected. The usage models need to be in place. I have to have a level of trust and comfort in the information I'm being provided. And then I have to be able to take action right then and there with it. That's the other piece that we're seeing. So just giving me information and serving it up is one thing. Giving me information, serving it up, and giving me the option to take action, then helping support me in taking that action immediately propels people towards the space faster. That was a long rant, Kyle. What do you think? Did I hit that one?

(Kyle Lagunas at 00:40:11) You landed the plane eventually. Okay?

(Allyn Bailey at 00:40:14) Thank you.

(Kyle Lagunas at 00:40:15) We're long-winded folks, Joel.

(Joel Beasley at 00:40:16) Eventually. I like that trust thing because the example I like to use with people is the Tesla car. If you've ever had the full self-driving in Tesla, there are the people that they'll see me, and I'll show up in the car, and they're like, "You let that thing drive? That's unbelievable. Why would you do that? This is so crazy." And I was like, "Well, it wasn't all in one day." It was, you know, I gave it a little test, let it do a mile, and I was like, "Okay." And then inevitably, some situation happened where it handled it equally as good as I would have. And then another time, it saw something I didn't even see and kind of saved my butt. And, you know, a couple different things happen, and then you just build this trust with it. And I think the same thing is true with the models. You know?

(Allyn Bailey at 00:41:04) To me, that's exactly why this is about momentum. Because what you just articulated was a basic psychological journey. As you build comfort, confidence, and a sense of trust over time, you can accelerate the pace at which you start moving forward, and you're willing to take those next jumps and those next jumps and those next jumps.

(Kyle Lagunas at 00:41:25) I think it's a great line of thought. I mean, momentum implicitly conveys competence, in my opinion. If we have been able to keep moving and keep—whether it doesn't mean every single pilot goes as planned. They're pilots for a reason. Right? We're seeing what works and what doesn't. But that we stay on the board means that we are committed. That's why your posture is important. Can I trust the posture of the business when they say what we're going to do with AI? Posture could be like, we're going to be human-centric, whatever. But, just like momentum is built, so is trust. And I think that's all the more reason why I just feel strongly about empowering HR to figure more of this out, to innovate with more confidence, because we work with some of the most sensitive AI use cases. And I have seen these use cases—at least in shiny demo environments. Of course, a vendor's sandbox is basically the same thing as the real enterprise. But some of this stuff could be really actually beneficial for us.

(Kyle Lagunas at 00:42:34) I know that there is—it remains a lot of discourse around, you know, how to—where robots are making decisions and where AI is impacting jobs that maybe it shouldn't. And there still is a lot of dust to settle. But at the core, the value proposition that these capabilities are offering to our colleagues in HR and talent are something that we've been desperate for, which is scale. And scaling effectiveness and not just efficiency, that's deeply compelling to me. And I really hope that we can find—I'm really hoping we can figure it out. I'm confident we're going to. Things are looking better than they were a year ago.

(Joel Beasley at 00:43:17) Mhmm. Yeah. And I don't imagine—I'll speak for the US mostly here because that's where all my experience has been. But I don't imagine that there's going to be too much proactive stuff as far as, specifically, you mentioned job automation. I think about that quite a bit because I'm a software engineer of twenty years building this stuff. So when I see it, it's like I've got a different perspective than people who are just technology savvy. And one of the things I'll tell you—what I wish would happen and what I think is going to happen.

(Kyle Lagunas at 00:43:47) Okay.

(Joel Beasley at 00:43:48) What I wish would happen is that we had some guardrails right now. Like, they proactively got together. They're like, "Hey, this is how we're going to handle this situation." Obviously, there's going to be rapid consolidation of industries overnight at a speed which we have never seen, and that is still yet to come, and it'll accelerate even greater than anything we've ever experienced in our lifetimes, which is going to create these dips and these issues that are going to be niche things. So you're going to wake up one day, and this entire market segment, whatever it may be, will be fully and completely automated, and it'll just happen. And then what's going to happen is there's going to be a lot of social unrest in that situation, and then we'll come up with some way to handle that. We are always—historically, look at seat belts, all of this stuff. We are always lagging behind the tragedy when it comes to policy. So policy lags tragedy all the time. And so when it comes to AI, I'm like, look. I don't know when it's going to happen, but it'll happen after there's a lot of tragedy in the marketplace.

(Allyn Bailey at 00:44:51) Yeah. It's not always the best thing to send it to a Gen Xer like me who immediately sees Terminator coming out of the woodwork right now. But, yes, you're absolutely right. Yeah.

(Joel Beasley at 00:45:00) Yeah. I see the—our generation sees it too. We're like, well, do something. Everyone's just pointing at everyone saying do something.

(Kyle Lagunas at 00:45:08) Yeah. That's right. But do it right. You know? Because I think that we have seen attempts at regulation that have been overly burdensome and impractical, and that's the challenge with just internal governance too. We do see that the federal government in the US is more inclined to let the private sector work some of this stuff out, which is interesting because there are industries like financial services that you would expect to be like, "Not till we have some clarity," but they're actually moving forward and saying—because they have established very clear governance. They are ready for an audit. They are already been data-driven. They've already had a lot of these capabilities that are making AI possible. I'm seeing them moving faster than others, faster in different ways. It's—I think it'll be a combination, Joel, of tragedy and positive ROI, like, proven ROI. You know?

(Allyn Bailey at 00:46:14) We're going to call those happy accidents. Right? In the human existence and relationship with technology, there is what we intended it to do and then what it ends up doing. And what it ends up doing ends up being some frame of, "Oh, look at that. That's cool. Didn't know that was going to happen, but, okay, that's kind of cool." And then we all act like that was what we intended, and then we'll move on and evolve it and kind of move it forward. And there'll be these moments where it does something and we go, "Oh, crapola. That is not what we intended to have happen," and we'll either figure out how to contain it. Right?

(Joel Beasley at 00:46:50) Absolutely. Yeah. And AI—I'm long-term optimist. I am long-term optimist. But I'm also not going to stick my head in the sand at the bloodbath that will be specific niche sectors really quickly. Right? And the positive to those moments is that humans are slow at implementation. So the technology is out here today and available to do unbelievable things. But as humans, we have to communicate, translate, project, plan, execute. And so there—I think it's almost a feature. I think the slowness of the humans' implementation ability is almost a—say it's like a safety feature that I think we're going to see play out. So I think that's happening. Yeah.

(Joel Beasley at 00:47:40) That's my positive for the short term is just human nature. My positive for the long term is also human nature because at the end of the day, Kyle and Allyn, what we do as people is we trade value within each other, to each other. Money is just the medium of the way we exchange value. Humans will not stop exchanging value. The things that they do to exchange value will change over time. If you just look past the hundred years—that's a constant change. And so that gives me optimism. The only part that I'm like, that's a real problem is the rate at which that change happens is faster than ever before.

(Kyle Lagunas at 00:48:21) Yep. I think we're going to have to put this into the model, Allyn. Value.

(Allyn Bailey at 00:48:26) I actually—you just—you got me on something. Yeah. Value. It's very connected to a lot of the conversations we're having around the future of work as well. Right? What is work? What is work? It's an exchange of value. How does that start to shift? You just got me, Joel. My whole brain just went, "Oh, I have—yes, Kyle. We have someplace to go here." It's very good.

(Kyle Lagunas at 00:48:44) Yeah.

(Joel Beasley at 00:48:45) Well, if you ever want to talk to me outside of the podcast, I'm more than happy to. I have thought about this in a very—I've spent hours just thinking about it for no particular reason. Just for me to understand where—well, I saw the deepfakes, Allyn. And this is what I do. And so how much time do I have? And I just went down this rabbit hole, and I was like, I'm trying to figure out how human behavior is going to change so that I can position myself and my family in the right spot so that we don't drown. Right?

(Allyn Bailey at 00:49:19) Yeah. We're all out here reading the tea leaves, Joel. Kind of like, I don't—you know? I'm fine. But you're absolutely right. It is right now—the moment to kind of sit back and go, "Okay. What will tomorrow look like?" And there's many paths it can take. There are many paths it can take, but there are the fundamental truths. How are we as humans? How do we interact as humans? What do we create value in? How do we build our own sense of identity? How does that connect to our sense of global connection? I know that gets a little woo-woo, but I think that's where we're—you know, we will figure those pieces out. We'll figure those pieces out. Right?

(Kyle Lagunas at 00:50:00) Yeah.

(Joel Beasley at 00:50:00) You're a little woo-woo. You have to be a little woo-woo to kind of see the future. Now, Allyn, SmartRecruiters, now SAP company, SAP unveiled their autonomous enterprise vision at Sapphire. Tell me about that.

(Alin Bailey at 00:50:14) Yeah. So here's what's exciting about this, the autonomous enterprise framework for us. And I laugh when I said we knew we were gonna— the reason I laugh a little bit is that we're adding more words into a space that we've all been trying to kind of navigate and struggle and put our arms around, and I get that. And autonomous is really this understanding that we are moving towards this space where what work is, how we perform work, that's gonna shift and change.

(Alin Bailey at 00:50:41) Technology is gonna be doing more and more of these bits and components, taking actions more independently within these guardrails and frames that get developed. And that's a whole bunch around this. You know, the SAP really great smart brains can talk about what that means in terms of data and data alignment and process alignment and all of those biggest components. When I take it down to a root level for us as a company, and we think about SmartRecruiters, we're really connected to this human capital management component, this larger story, right? How do we connect humans to work?

(Alin Bailey at 00:51:13) And hiring, which is what we do at SmartRecruiters, is the front door to all of that. It is the front door to how we create the data pathways that we start to get so we start to understand who people are. It is the front door in which we're able to look at the information coming in and to be able to apply it appropriately throughout the lifecycle of that person's relationship with the company. You know, we're way past the days— not way past. I believe we're getting way past the days where it was a function of send in your resume, and it just sits in this database. It's not connected to anything over here, and we're never gonna go touch it again. And that's always been a challenge because it's a very static document. Right? I mean, we learn, we change, we evolve so quickly.

(Alin Bailey at 00:52:01) So to really be— for SmartRecruiters to really be connected to this larger story, which is how do you start connecting and understanding about what both what a company needs, that kind of workforce planning data and information, and applying it successfully to being able to get the right people in the door and connecting them to the right opportunities, and then helping them understand where their value is in that place. That's a huge opportunity. So autonomous enterprise starts getting us down that path because it connects the data. At the end of the day, this is a data story, and the technology is an overlay to help us with decision support.

(Joel Beasley at 00:52:39) I love it. Kyle, Alin, we made a podcast. How do you feel? Awesome.

(Kyle Lagunas at 00:52:44) Fabulous, Joel. It's always a pleasure to sit down with Alin Bailey. She's brilliant.

(Joel Beasley at 00:52:50) Thank you so much for listening. And if you found this episode useful, please share it with a friend or colleague who you think would get value from it. And if you have topics you'd like to hear discussed on the podcast, either add me on LinkedIn or send me an email, [email protected]. Every time I get an email or LinkedIn message, it absolutely makes my day and inspires me to keep going.