AI in Hiring: Trust, Safety, and Why Compliance Freaks Out

Forget “digital transformation”—this is full-on AI combustion. Allyn Bailey from SmartRecruiters joins Chad & Cheese to explain why breaking your ATS is a feature, not a bug, and why TA pros secretly need therapy every time someone says “compliance.”

Inside this episode:

• Why legacy HR tech is basically a zombie apocalypse in slow motion

• AI safety councils that sound more like Fight Club (but with lawyers)

• Global wildcards like Kenya proving Silicon Valley ain’t the only AI show in town

• And the sad reality that scheduling is still the hill recruiting tech dies on

Season 3 isn’t here to whisper sweet nothings about AI. It’s here to rattle cages, spill beer, and roast bad tech until it cries.

Buckle up, kids. The Kraken has been released. 🐙

PODCAST TRANSCRIPTION

Joel Cheesman: All right, let's do this. We are the Chad and Cheese podcast. I'm your co-host, Joel Cheesman. Joined as always, Chad Sowash is in the shotgun seat. And this is the Sessions AI frontline series. And we welcome Allyn Bailey, Senior Director of Communications at SmartRecruiters. Allyn, welcome to HR's Most Dangerous Podcast.

Allyn Bailey: I am so excited to be here. I feel like I haven't seen you guys in so long.

Chad Sowash: What?

Allyn Bailey: Well, I see you all the time.

Chad Sowash: We were just in Madrid together.

Allyn Bailey: I haven't talked to you on a couch in a very long time.

Joel Cheesman: She's guilting us already. Already the guilt trip.

Chad Sowash: I went to Madrid for her.

Allyn Bailey: You did, you did. But I don't talk to you guys very much anymore. I feel like I'm not in the talking space anymore.

Chad Sowash: I blame you. I blame you.

Joel Cheesman: Let's talk now. Let's talk now.

Allyn Bailey: Okay, we'll talk now.

Joel Cheesman: But before we get into that, a lot of our viewers won't know you. So give us the elevator pitch.

Allyn Bailey: Okay. So I've been in the talent acquisition space now for almost a decade, which I cannot believe. It's probably longer than a decade. I don't want to add it all up. I came from the practitioner side of the house, was at Intel, worked there for a very long time doing talent strategy and talent design work. Then was off doing my own spiel for a while, doing some consulting, working with all sorts of different types of companies. And then landed here at SmartRecruiters, where I've really been kind of working on the intersection between what's really happening in the practitioner side of the house and what are we building from a tech perspective and got deep in the weeds.

Chad Sowash: So we see a good amount of practitioners come out of big names like Intel, and they go into a vendor, right? And they usually flame out because it is a different world. It is a different world. You've not done that. I mean, you've been in, this is... I mean, you've absorbed this. What's the difference? Because there's a huge difference, right? What's made it a little bit more sustainable for you as not just a job, but really, I mean, something that you put your life into from a career standpoint?

Allyn Bailey: Well, I think even when I was on the corporate side of the house, right, kind of living in that practitioner side of the house and the corporate world, I was never one that was kind of just doing one thing all the time.

Chad Sowash: And you were very much in the tech, too.

Allyn Bailey: That's right. I was very much deep into the tech. I was always shifting and trying new things. I came from the user design and the user experience side of the house. That was what I was, my focus was at. And so I think now working at a tech company in a much more specific frame, it allows me the opportunity to do lots of different things, to pivot around very quickly, which works with my style. So it's actually probably even a better fit than the corporate world was and kind of working in a very structured setting.

Joel Cheesman: I'm probably dating myself when I say she's the EF Hutton of her company. When she speaks, people listen because I don't think a lot of vendors hire people from the actual customer side and more should do it. I assume that you've been a real benefit...

Chad Sowash: They try.

Joel Cheesman: Of being on the vendor side. Talk about that.

Allyn Bailey: Yeah. Well, I mean, I think I've always, one of the things, I was a big proponent of SmartRecruiters before I even came over to this side of the house. That was part of that was part of the deal for me. Right? I wasn't going to work for anybody. I think people know that about me. I am pretty frank. I say what I'm thinking.

Joel Cheesman: No.

Allyn Bailey: I know, right? So that's been, just is who I am. I'm not going to not do that. And I think SmartRecruiters has been a company that has been very welcoming of that. They've asked me to say what I think. They've asked me to be part of that conversation. And so, yeah, that's been that's been a big benefit, I think, for both of us, right? I started when I first came over, started working on the consulting side of the house with them and working with companies to help them understand how to use the tech more effectively and then quickly pivoted into actually doing services and implementation. And that was all because we quickly realized it wasn't just about kind of quick fix consulting solutions. There was a deeper conversation we need to have about how to actually implement technology. Having been on that side of the house and driving that conversation, SmartRecruiters was very open to me having that dialogue. And then that transitioned into, okay, well, here's what people are really trying to do. So let's have a conversation about what type of tech we're building.

Chad Sowash: So SmartRecruiters is also open to a very large pivot. Now, we take a look at the model was built literally on legacy process. Right? And you guys have well over 2000 customers. It's hard, which is why we see a lot of applicant tracking systems, just a lot of platforms. They literally, and I always use the Taleo instance, is that literally they just atrophy and die, you know. But you guys made a huge decision already being in a specific architecture, already having a portfolio of 2000 companies and customers to be able to make that pivot. That is risky as hell. As a comms director, it's got to be somewhat exciting, but also it's risky. Tell us about that change. Why was the change made necessary and how exciting and or scary was it?

Allyn Bailey: So it wasn't scary, at least to me, because I like to shake shit up. So I'm going to do that anyway. Here's the deal. I think it was and has been a process of really understanding what people were trying to accomplish with the tech and realizing that we are, because of what technology is now capable of doing, we're in a once-in-a-generation moment where we can actually do stuff with the technology that we've been talking about as a TA profession, as a human resources profession, literally for decades, right? We've been saying, well, I want to talk about great candidate experience, or we want to talk about being able to engage hiring managers so they can actually understand what and how they're hiring, provide them insights that are at a moment's notice. These were all things that we were creating technology for and leveraging technology to do, but we had all these workarounds built into it, and it was complex and bloated and frustrating to use. And then at the end of the day, people just weren't using it because you start adding too many components to it, people just go back to the basics and say, whatever, right? I'm just going to do it the way I've always done it and I'm not going to shift and change.

Chad Sowash: It's almost impossible to navigate when you have feature bloat at that point.

Allyn Bailey: That's right. You absolutely can't. So I think having a strong leader come in, and that's, you know, we brought Rebecca in, or, you know, Rebecca came in back into the fold. She had a strong vision, but also a strong enough presence to be able to say, in a sense of kind of where we were in this moment to say, it's okay. I'm going to trust all of you if you trust me. Let's break it and fix it. Let's say, if we didn't do it that way, what would we do? That's a hard question for anybody to, one, ask or to be really willing to accept the answer to. And then when we all started talking about it and saying, well, what would we do, we realized, oh, shoot, we'd do something completely different. Let's do it. So on that level, that part was easy. I think what you're pointing to is a second part, which is really complicated. I can tell you part of these conversations that we're having now and the conversations we're having with our All In for AI series across all sorts of different groupings and with customers and with people across the industry who are in this space...

Allyn Bailey: It is about that second part of the conversation, which is if I am already a customer, if I'm already embedded in how we do business today, convincing you, dear customer, convincing you, TA practitioner, that the way we're doing it today may not be the best approach and we can do it a different way and here's the new technology to do it. That is a whole change management process that is difficult, right? That's a paradigm shift from the normal vendor strategy, which is, dear customer, tell me what you want and I will build it for you as you have asked. Instead, we are looking at it and saying, listen, we've been watching for a long time. We understand where the industry is going, but we also understand what needs to happen in this space. We understand what can be helpful for you. We're going to propose and give you some new options and new ways to do this and it may not be how you've done it before and it may put you in some uncomfortable conversations in your own teams.

Chad Sowash: Well, I feel like the agents, though, makes it a little bit easier because everybody wants to feel special with their own process.

Allyn Bailey: Absolutely.

Chad Sowash: And now they can feel special with their own agents and being able to actually adopt those own agents that could be more standardized to a better process methodology that you guys have actually seen over the years, but then still it's that customized feel of having an agent, much like ChatGPT feels like. For anybody who uses it.

Allyn Bailey: Yeah, of course. Right, right. And the more you start to leverage it and you start to get comfortable with it, companies are starting to realize they can create experiences that are unique to them, both inside and outside their company, leveraging these tools.

Joel Cheesman: One of the things that I've found really fascinating is when you have technology in our space that has sort of gotten on in the years and got some age under it.

Chad Sowash: It's old.

Joel Cheesman: It tends to entropy, excitement leaves, the energy sort of goes somewhere else, and you guys somehow harnessed a win attitude. We're playing to win, we're not playing just not to lose. And I'm curious, where did that come from? Was it the vision? Was it a concerted effort to say, you know what, look, we've been around a while, but we're going to shake it up and we're going to win again? Because it's permeated not only in the company that I see, but with the customers and people taking a second look at you. Where did that come from? Was that Rebecca on down? Was it the whole organization? How did that come about?

Chad Sowash: Everybody hates the prevent defense, by the way. [laughter]

Joel Cheesman: The Four Corners, remember that?

Chad Sowash: Yeah, yeah, yeah, yeah.

Allyn Bailey: Oh, yeah, right? Yeah, exactly. It's a great question. I think it is a perfect storm. It's a perfect storm of moments, right? The right team, right? And it's an interesting team. People would say, you look at us and say, we're pivoting, we're looking at doing things in a different way, we're looking at things in a new way. You would think it's a whole new group of people who have shown up on the plate. That's not the case. Our senior VP of engineering has been with the company over a decade, almost 12, 13 years. He built the original components of the platform, right? Rebecca is a returner into this space, right? Kind of the old product leader coming in and taking over the helm as the CEO. Our tenure across our organization is vast, right? We have everything from new people in the door to people who have been there five, 10, 15 years, right? So the energy really comes from an alignment around the vision. When we start to see it work, and we're starting to see it work, we're seeing customers who are in our design labs who are actually trying this stuff out and are getting excited about what they're capable of doing.

Allyn Bailey: It creates momentum and excitement, right? And we have more and more of that dialogue and conversation around it. We start to see it working with these use cases. And then we start to operate in a new way that says, what if? When you start asking people what if and you release the handcuffs and say, you've been doing this for a while. You maybe have some ideas that we've been talking about in the back rooms and behind closed doors and in different alleyways. And we have all those dialogues, right? We sit in the conferences and all these conferences, right? Everybody says, listen, you can go sit in all the sessions, which are fabulous and are great, but where do the real conversations happen? They happen in the hallways. They happen in the booths. They happen around the corner.

Joel Cheesman: In the booths. After hours.

Allyn Bailey: After hours, right? I think our company has started to build this energy around saying, if we took all those after hour conversations, what could we do? That gets exciting. People get energized about those. So vision, a little bit of seeing it actually come to fruition, and that's exciting. And then really great leadership coming in and saying, I release the handcuffs. What would you do?

Joel Cheesman: "Release the Kraken." I love "release the Kraken".

Allyn Bailey: "Release the kraken." That's a great way.

Chad Sowash: So the AI council in itself, talk a little bit about that and, I mean, why AI safety? I mean, obviously adoption is happening much faster than I think most of us thought it would happen because of OpenAI and people really using it on their phones every single day. So it's not, I don't think it's as hard for us to be able to push for adoption. What was the safety, the risk part of that that you guys wanted to try to wipe out of there?

Allyn Bailey: Right. Well, first off, I don't think we can wipe it out, and that's been part of our honest dialogue with both our customers and, by the way, just general people within the industry. AI is really cool. It's going to do some really neat stuff, right? We're really excited about it. But it also has the potential to do some really scary things because at the end of the day, we're looking at all this great tech, but all this great tech is based off of what we've been doing already. When we talk about large language models, what are they learning off of? They're learning off of data and processes and systems and behaviors that's biased because we as humans, I hate to tell everybody, we are really not great. We are horribly biased, right? We just, by our nature. Technology is going to be biased too because technology is built off of how we operate and think. So if that's the case, you add that challenge in. Now you also add in the challenge of we're a global company. Our footprint is in over 100 countries, right? We are all over the place.

Allyn Bailey: Every country today, every state today, every region today is starting to build its own set of AI guidelines, laws, regulations. That's very complex, very hard thing to navigate. At its root, talent acquisition, HR in general, has been based on a compliance methodology, right? So we are a group of people who like to say that we're really cutting edge. When I say group of people, I'm talking about talent acquisition, but we're really deep down a whole bunch of rule followers. So it's freaky, right? It's scary. So we've got this challenge. As a comms director, I look at it and say, okay, you've got a lot of people who are interested in having backroom conversations about AI, but they're also very scared, very worried, unsure of how this is all going to work. Rebecca says this really well. You only get one chance to put AI into your organization or to start to implement it without it leaving some sort of taste in everybody's mouth. It feels good and it feels comfortable and I trust it or something goes wrong. It only needs to go wrong once for everybody to freak out about it. So if that's the case, we need to have open conversations about what are the risks? What are the challenges? How do we solve these problems?

Allyn Bailey: We don't know all the answers. We didn't want to go into a black box and create all the answers ourselves. So we invited customers in, customers from around the world, customers from our big companies, from our small companies. We invited practitioners who are not even customers to come in. We invited, we have some legislators in the conversation.

Chad Sowash: Smart.

Allyn Bailey: And we're having open conversations. Again, we're doing it behind closed doors. We told, listen, we're not going to talk about it. We're not going to have you come out and tell everybody all the great legislation that we're having conversations about. We're not going to have a conversation about the choices that we're making, but we want you to have a safe space to have a real conversation about how do we trust this and what do we need to know as people building this tech that's going to make it valuable, trustworthy, and usable for you. And usable is the key because...

Chad Sowash: Are these people piloting?

Allyn Bailey: Some of them are piloting.

Chad Sowash: Okay, because that would be a nice little bait to come into the, you know, I get to use the new stuff.

Allyn Bailey: Well, so to get to the new stuff, the Trust and Safety Council, that's really about, let's talk about the challenges here and let's debate how we solve them. We have a second group of people who are a series of customers. They are part of our design group and our design lab. You're going to talk to Matt a little bit later. He's part of that team from Domino's. Listen, they are using the early stuff. They are helping us design it from the root up. We're having conversations about how it works inside their businesses, and then we're learning from them and helping expand that out to everybody else. So those people, those people in the design groups, yeah, they're getting early access.

Joel Cheesman: You've always been a big picture thinker, and I'm curious your thoughts on who's embracing the technology, who's a little bit more timid around it, whether that's small versus big co, parts of the world that are sort of embracing it more than others, businesses maybe that are embracing it more than others. What are you seeing from sort of a macro perspective on the adoption of AI?

Allyn Bailey: It's a great question. I think there are two things that I think are unique in this AI adoption space. I think whether it is regions, countries, companies, or kind of just groups, right, that are most likely to leverage AI today, they are the spaces where they have less governance and traditional processes that are embedded so deeply. We see that even just in the evolution of AI in general, the best AI evolutions right now from an AI explorer perspective are coming from countries and spaces most of us aren't even paying attention to. The work going on in Kenya, fascinating. The work they're doing, you know, in some of the former Eastern European countries, huge, right? So that also translates to who do we see in terms of customers and potential customers who are raising their hand and saying, I'm willing to try it. Let me figure out how to do this. They're the people who have less repetitive process they have used for long periods of time that they have to break before they do something new. Because the thing that we're not always talking about yet, is the tech is cool, but the tech isn't going to work on your old processes and with your old roles and with your old systems.

Allyn Bailey: And so if I don't have a lot of technical debt or process debt to navigate, it's easier for me to adopt. The other big adopters, interestingly enough, they're not my TA and CHRO teams, right? It is mandates coming from CEOs, CFOs, CTOs. And the challenge there is they don't always know what's possible and what isn't possible. They just want it to happen.

Chad Sowash: Is this productivity and efficiency? Is that...

[overlapping conversation]

Allyn Bailey: Productivity, efficiency, and modernization. I would say modernization of their businesses and a belief system. I don't want to say it's hype. They are listening to the entire world out there telling them that business is changing. You need to be on this new technology bandwagon. You need to understand and have AI operating within your company. AI-first companies are going to be the ones that win. And so they are telling all of their leadership on down the path, go make that happen. And there's a challenge there, not all of those business areas are ready to go.

Chad Sowash: So when it comes to being able to AI your process, you're not going to go from start to finish, right? So for companies, especially that have been piloting, that you've been talking about, they're probably going to pilot a chunk, a task, a specific task that takes a lot of time and it demonstrates to the C-suite, hey, look, we're making big progress, but we've only done this, right? What is that task that you guys are seeing right now for these teams?

Allyn Bailey: You know what it is?

Chad Sowash: It's got to be something boring.

Allyn Bailey: Scheduling.

Chad Sowash: Yes?

Allyn Bailey: Scheduling, which by the way, drives me crazy because I'm like, this is seriously the biggest problem we have? But honestly, it is the thing that people get most excited about solving because it takes time.

Chad Sowash: 'Çause it sucks.

Allyn Bailey: It sucks. It is horribly complicated. I got too many calendars, etcetera. And I think, this is a great example of approaching the problem differently. So scheduling is the issue. And for most people, it is because it is an efficiency problem. It's complicated and it's hard to do just to get calendars to mesh up and etcetera, right? We actually took a second lens onto it and said, but okay, let's assume I can solve that because theoretically we can and we're already solving that, right?

Allyn Bailey: What then becomes the next root problem? The next root problem that AI actually helps us solve is, who should be interviewing? How do I determine based on the role that I'm leveraging or looking for, or based on the type of individuals that I'm looking to interview, and maybe that differs for each role depending on the applicant, etcetera, who the best person is to pull into that interview cycle. How many times do we have people sitting there trying to figure out who should be on the interview team? What are the questions they should be asking? How do I then assess the information they brought in some sort of simplified way so that everybody has a even playing card to look at? Those are problems AI can now solve. So scheduling is the root issue, but we can actually go at it deeper now and provide intelligence and insight that allows it to be an even better process. To me, that's the exciting piece.

Joel Cheesman: No one seems to want to stay in their lane these days. Today's chatbot is tomorrow's ATS. You know, today's job board is tomorrow's scheduling solution, etcetera. How do you guys look at that from a competitive standpoint, from a product perspective? I know Winston is a product that you've released lately or recently. How do you think about it competitively when it seems like every month there's a new ATS that wasn't an ATS before, but everyone's trying to be everything to everybody? How do you look at the competitive landscape from that lens?

Allyn Bailey: That's a great question. Let me tell you what I think. Well, one, if they're walking out today, cramming a new ATS, we'd tell you, well, that was so 10 years ago. That was a mistake. We're already past that. Yeah. I think competitively, the challenge here is focusing on the use cases that the tech is trying to solve rather than trying to define or to over-scope what something's going to deliver. Great example. We know this. If everybody's going to raise their hand today inside that conference hall and say, what is your biggest challenge, for example, with HCMs, right? What are they going to say? They're going to say, well, they try and do everything. Nobody gets the right investment into any particular component of the product, whether it be...

Chad Sowash: They do nothing well.

Allyn Bailey: And they do nothing well, right? So we all know that's the case. Why did we end up there? Well, there's a lot of reasons, right? It has to do with how CFOs want to buy, the idea that I want to have a simplified tech stack, all of those reasons. But we know from watching that, that that's a mistake. It's a mistake to try and do everything and do nothing well. So from our perspective, and as we look out there into the competitive framework, we say, listen, there are two ways to approach this. There is end-to-end hiring solutions built for hiring use cases. We think we're the only ones out there. I can have that conversation a whole 'nother time. And then there are best in breed point solutions that can be in space, right? Like maybe I have a really great video interviewing component that can be added on or tapped on or etcetera. I think our approach is to look at this and say, you can be an HCM and try and hit the full suite end-to-end. You can be a hiring end-to-end platform and then understand how you're going to approach that from a very specific use case perspective to say, how do I look at four big decisions that have to get made, right? Who do I need to hire? What type of skills do I need, right? Who am I going to bring in to assess? Who am I going to then decide is the right choice? And then how am I going to get them onboarded?

Allyn Bailey: End of the deal, right? If we can handle that, we do it well. We focus in on the people who are working in that space.

Joel Cheesman: All right, we will end it there. Allyn, thanks for hanging out with us. And this has been the Chad & Cheese Podcast. I'm Joel Cheesman. He's Chad Sowash. And this has been the Sessions AI Frontline Series. We out.

Chad Sowash: We out.