Recruiting Future is a podcast that helps Talent Acquisition teams drive measurable impact by developing their strategic capability in Foresight, Influence, Talent, and Technology.
This episode is about Technology.
The AI transformation in recruiting presents both unprecedented opportunities and significant threats. It’s a complex situation where the same technology that promises to solve long-standing recruiting challenges is also creating new ones. The use of AI by candidates is driving up application numbers, and Gartner predicts that by 2028, one in four job applications will be fake or fraudulent.
So, how can employers ensure they are reaping the benefits of using AI in the hiring process while effectively mitigating the risks?
My guest this week is Meredith Johnson, Chief Product Officer at Greenhouse. In our conversation, Meredith shares her insights on building ethical AI systems that enhance rather than replace human judgment. We discuss creating fraud detection capabilities that protect hiring quality and designing embedded AI experiences that maintain the transparency and trust that great hiring requires.
In the interview, we discuss:
• Maximizing AI benefits while reducing risks
• Building C-suite support for AI investments
• Embedding AI within existing workflows
• Detecting and filtering fraudulent applications
• Talent filtering, interview question generation, and sentiment analysis
• Maintaining ethics and reducing bias
• Building transparency into the candidate experience
• What does the future of TA look like?
Follow this podcast on Apple Podcasts.
Follow this podcast on Spotify.
00:00
Matt Alder
For many employers, AI is currently a double edged sword. Its potential to drive better hiring outcomes and radically improve the candidate experience is enormous. However, at the same time it’s increasing application volumes and massively expanding the opportunity for fraud. So how do you benefit from the positives while defending against the negatives? Keep listening to find out. Support for this podcast comes from Greenhouse. Greenhouse is the only hiring software you’ll ever need from outreach to offer. Greenhouse helps companies get measurably better at hiring with smarter, more efficient solutions powered by built in AI. With Greenhouse AI, you can generate stronger candidate pools faster and source high quality talent with more precision. Streamline the interview process with automation tools and make faster, more confident hiring decisions with AI powered reporting.
01:03
Matt Alder
Greenhouse has helped over 7,500 customers across diverse industry verticals from early stage to enterprise become great at hiring, including companies like Airbnb, HubSpot, Lyft, SeatGeek, HelloFresh and DoorDash. If you’re ready to put the power of AI into the hands of your hiring team, you can visit greenhouse.com to learn more.
Matt Alder
Hi there. Welcome to episode 710 of Recruiting Future with me, Matt Alder. Recruiting Future is a podcast that helps talent acquisition teams drive measurable impact by developing their strategic capability in foresight, influence, talent and technology. This episode is about technology. The AI transformation in recruiting presents both unprecedented opportunities and significant threats. It’s a complex situation where the same technology that promises to solve long standing recruiting challenges is also creating new ones.
02:28
Matt Alder
The use of AI by candidates is driving up application numbers and Gartner predicts that by 2028, as many as one in four job applications will be fake or fraudulent. So how can employers make sure that they’re reaping the benefits of using AI in the hiring process and while effectively mitigating the risks? My guest this week is Meredith Johnson, Chief Product Officer at Greenhouse. In our conversation, Meredith shares her insights on building ethical AI systems that enhance rather than replace human judgment. We discuss creating fraud detection capabilities that protect the quality of hire and designing embedded AI experiences that maintain the transparency and trust that great hiring requires. Hi Meredith, and welcome to the podcast.
03:20
Meredith Johnson
Thank you. Glad to be here.
03:21
Matt Alder
An absolute pleasure to have you on the show. Please, could you introduce yourself and tell everyone what you do?
03:27
Meredith Johnson
Absolutely. Well, thanks again for having me. I’m Meredith Johnson. I’m the Chief Product Officer at Greenhouse. And that means that I Lead the product strategy, design, delivery, and I oversee all of product management to make sure that we bring really great value to all of our customers, especially in the TA space.
03:41
Matt Alder
Really interested in your sort of view on the market. Obviously as a product professional, I know this is something that you’ll be looking at all the time. What’s the sort of state of recruiting right now? Where do you think TA leaders need the most help?
03:55
Meredith Johnson
Yeah, well, listen, this is a really important question because so much has changed over the last, say five or seven years, but I would say really just over the past two years the biggest thing that we’re seeing is with large language models and AI tools have, you know, as they’ve taken hold, that type of funnel, that number of applicants is really challenging. And that’s one of the biggest challenges that’s facing TA in recruiting teams today. If you think about some of the statistics, the number of applications has doubled just in that 24 month period. And that means that the number of applications per recruiter has almost tripled. Right. So that influx is really leading to two things that I’m seeing and hearing from our customers. One, terrible experiences for our TA teams and also not great experiences for candidates as well.
04:39
Meredith Johnson
So recruiters are overwhelmed, candidate experience suffers. And it’s a real challenge to kind of, you know, filter through that volume.
04:47
Matt Alder
No, absolutely. And I’m sure that there’ll be things that we sort of talk about as we, as we move through the conversation because they are, they’re just massive problems and they’re issues that are only going to get worse and become more important, aren’t they?
05:00
Meredith Johnson
Yes, that’s for sure.
05:01
Matt Alder
From a technology perspective, I kind of just want to start really at the beginning because lots of people listening will be looking at their technology stack at the moment, thinking about AI, thinking about how they can really kind of embrace technology to help with some of the challenges that they have and move forward. One of the biggest challenges around that is getting the budget and resources to implement new tools or enhance existing tools. What would you advise people in terms of getting sort of C suite buy in to sort of buy AI technology or enhance AI technology within their technology stack?
05:35
Meredith Johnson
Yeah, you know what, this is a really important question. Especially as you know, within different sectors of our industry and different size and complexities of company, there’s varying degrees of reticence to not only be able to invest in tools that really matter and can help teams, but to adopt new technologies like AI that can actually help drive not only productivity and efficiency, but efficacy of getting to those great hires. So it’s a really good question. I think, you know, the advice that I give customers and a lot of things that we talk about is having teams really focus in on the strategic business outcomes, right? So we just talked about some of the challenges that, you know, this massive amount of applications creates, right? There’s too many candidates. We haven’t even talked about spam infrastructure which we can get into later.
06:19
Meredith Johnson
And it can be a really challenging, scary thing. And so I think helping teams connect with their executives on what are the strategic business outcomes that we need to achieve, right? How will we track them, how will we measure them? What’s the ROI is really critical because then you can start to have a more strategic conversation about this vendor, this partner, this technology is going to help us get to those outcomes. And here’s why, and here’s why we believe it. At Greenhouse, we spend a lot of time with our teams and our customers trying to tie out that really important convers, not only for the users and the buyers, but as you asked, that important kind of C suite executive level that’s often not close to the day to day operations.
06:57
Meredith Johnson
And I think, you know, focusing on that alignment is really kind of the equalizer that can help bring everybody along regardless of level.
07:04
Matt Alder
And also I think one of the other issues for people is the integration of all of this in terms of what they’re doing. Because some of these issues that we’re talking about need to be, you know, we need some pretty quick solutions to them before recruiting just implodes on itself. How can people introduce AI into their exist workflows without sort of massively overhauling what they’re currently doing?
07:25
Meredith Johnson
Yeah, this is great. Well, you know what, I love this question because it really speaks to the way that Greenhouse has gone about thinking about creating embedded AI experiences right, where people are working as we build toward that future of having a more agentic product in and of itself. So I can give you a couple examples and I think this is really meaningful for teams, right? You want to feel comfortable, you want to trust, you want to be able to test and interact with these new technologies and you want them right where you are, right? You don’t want to go out of the system and do something in ChatGPT and have to copy it back in. And so we’ve been very purposeful and very deliberate about right within the workflow whether teams are setting up a job.
08:02
Meredith Johnson
They are Sourcing that, you know, that large funnel that we just talked about, they’re working through the interview process. We have new capabilities, almost 20 that are AI supported and AI assisted that are embedded right into that workflow for our customers. So for example, in Greenhouse now you can use AI to automatically generate job descriptions. So literally within seconds better than we could write them ourselves. It’s taking information right from your system and it’s creating that job description that is tailor made to the job that you’re trying to hire for with focused attributes. And that’s really important. Right. So that’s all about time saving, but it’s also about efficacy. Right. We want to have a JD that says what we needed to say to find that right talent. Right. Other things that we’ve done is started to trigger with AI sentiment analysis.
08:48
Meredith Johnson
So looking at all of the interview notes, we’re synthesizing and giving signals for sentiment for those applicants. So it might tell you that hey, this person’s really interested or this one might not be right. They’ve mentioned some timing concerns. Right. And it gives more time to the humans that have to make the decisions because you know, back strategic time because it’s giving you those signals. One other thing that I actually love, and it’s our most favorite feature, our most widely used FEature is auto generated interview questions. So if you think about that really purposeful job description that you’ve spent time creating, hopefully even using some of the AI assistance to build that out, it’s going to automatically suggest interview questions based on that job description based on key attributes.
09:32
Meredith Johnson
And it’s been really exciting to see our customers interact with it because almost 85% of them have generated questions and accepted the suggestions without edit, which is really neat. So again, it’s just, it’s about putting it right where you are, making it seem simple, making the UI super intuitive and also giving optionality to adjust or accept or reject the support that you’re getting from those new technologies.
09:55
Matt Alder
And in terms of those volumes of applications, I mean, what can help there with these kind of tools?
10:00
Meredith Johnson
Yeah, well, I mean, I think first off, you know, and this is actually an exciting time, there’s so many tools out there. You know, one of the things that we’re focused on is a couple things helping filter out or filter in, I should say, to a prioritized inbox, people who really, truly have a skill match. So Greenhouse is going to be releasing some really great, important functionality that’s going to help filter and build that prioritized inbox for you also, one of the things we haven’t started to talk about is in that vast pool of applicants, right there is this increasing volume of misrepresentation spamming. So it might not be bad people, but people who really don’t have a high affinity to your company or to your job, they’re just trying to put out as many applications as possible.
10:44
Meredith Johnson
I mentioned misrepresentation can kind of filter its way in there too, and then outright fraud. So it’s a really kind of big burden for RTA teams. But also it can be really scary and risky for companies, especially if they accidentally let someone in who wants to do harm to the company. And so the solutions that we’re providing help with not only filtering and talent matching, but also giving indications of potential fraud and so that we can kind of exclude ineligible applicants.
11:15
Matt Alder
There’s obviously lots in the news at the moment about discrimination and issues kind of around that with technology, how can recruiters sort of be certain that they’re not missing specific characteristics when candidates are being evaluated or that the system isn’t doing things that they wouldn’t do As a recruiter, how does that all work?
11:33
Meredith Johnson
Yeah, it’s really important. And it’s really important depending on where you live. Right. Because we’ve got varying levels of law that help protect individuals not only in terms of private data, but also how they are treated. So this is a really important question. I’m glad you asked it. You know, I think number one is making sure that, you know, back to one of your first questions. You are selecting and working with partners who you really trust, who are really transparent about how they handle data, how they think about ethics, how they think about integrity. So that’s super important. Right. And at Greenhouse, we have a very strict ethical standard and ethical committee that for particular for our AI features that everything we developed is run through that committee. Everything that we do around AI is published.
12:19
Meredith Johnson
So we are 100% transparent on what the features are, how they work. And actually our customers have optionality for which ones they want to turn on or not. Most over 75% turn them all on. And we’re getting some really great adoption. But there is some optionality there. I think the other thing is, within those solutions, you want to be looking for the ability to still maintain control and decision. And so that’s what we do. We are assisting teams, but the human makes the decision, and that’s really important. And you want to give some availability within the solution to check your work. Essentially, for example, one of the summerative Solutions that we’ve rolled out is scorecard summaries.
12:59
Meredith Johnson
This is really cool because it takes information from the scorecard, summarizes it up across the interview team and says, hey, here’s some points where you all are aligned and maybe points where you’re not. So it can help that team more quickly and strategically decide what they need to talk about regarding that candidate. But what’s really cool is it links and cites, all of those summaries. So if you want to click in and double check or understand exactly what was said or where it’s pulling that information, you can. So you have that confidence as you’re using the system that not only do you maintain the decision making, but you can also check what it’s citing. I think that’s really important. And I think with regard to equity, Greenhouse remains committed to building tools regardless of AI. That is, that supports equity. Right.
13:41
Meredith Johnson
And that supports the reduction of bias. So when we think about the time and energy we put into the prompt engineering behind some of these tools, they are purpose built and designed to eradicate and reduce bias. And that’s really important and it’s really cool too.
13:54
Matt Alder
And I just want to sort of reflect back to what you were saying about the, you know, the fraud and the privacy and all that kind of stuff. I mean, what else do you do to ensure accuracy and also kind of maintain data privacy? Because that is obviously a massive issue for lots of organizations at the moment.
14:09
Meredith Johnson
Yeah, absolutely. Well, very similar to our ethical standards and the committee that we’ve established as a kind of sister organization to all of the things that we’re doing within product development. We have a very public and transparent security and privacy policy as well that adheres to all legal and regulatory regulations, GDPR and everything you would assume. We’re also in the process of gaining our ISO 42001 certification, which aligns to not only how we use AI responsibly and ethically, but it’s also in compliance with all of those regulatory requirements. So again, I think, you know, I kind of put it in the buckets, as you’ve asked here, as ethics, as security, as privacy. And it’s super important. And so we adhere to those standards, we’re very transparent and we publish everything that we do, including all of our certifications.
14:59
Matt Alder
And are there any sort of specific questions around that you would recommend TA leaders ask vendors? We’re talking about lots of things kind of all sort of put together here and it can be quite confusing. Are there any sort of really important questions that People should be asking their vendor.
15:11
Meredith Johnson
Yeah, I think so. You know, and a few places I would start is, you know, one, I would be asking what are your ethical principles for how you develop software, including, you know, components that are AI supported or leverage large language models? I think that’s really important. So ask for it. And it should be a well documented, you know, transparent, you know, artifact that you can get to, ideally even just on the web. So I think you need to ask for that. I think you need to ask for what are your security and privacy standards? And that also should be very transparent and very easy to find. It actually should be something that’s probably, you don’t even have to ask for. It should be pretty easy to find. But I think that’s really important.
15:47
Meredith Johnson
I think within that you want to be looking for what certifications and regulatory compliance measures have been met that also should be really readily available within those policies and procedure documents. And then I think when it comes to particular functionality, really asking for transparency around, great. How is that made? Are you using OpenAI? Are you using a custom homegrown large language model? Are using a combination of both? Can I turn it on? Can I turn it off? When I work through the UI and actually interact with the solutions or it’s demoed to you, can you observe the ability to check or to refresh or to edit what is being returned to you? Then you always want to be looking for anything that looks like it is bringing a, you know, doing the decisioning, right? We want decision support, we don’t want decisioning.
16:39
Meredith Johnson
And so I think those are some of the things I’d be asking and looking for as I’m evaluating vendors and also starting to experience and demo new tools that are available.
16:49
Matt Alder
And from the candidate perspective, I think there are huge amounts of things here that should make life better for candidates. But how do we kind of position that how transparent should organizations be about the technology they’re using? What’s important from the candidate perspective here?
17:05
Meredith Johnson
You know what, I was talking with someone the other day and they actually, they asked the question, you know, do candidates know that AI is being used? And I kind of smiled because, you know, I’ll get back to your question of like, how transparent should we be? But they now, right, I would argue individuals, all of us in our personal lives and certainly in the discussion here that we’re having about people applying for roles are way more advanced, way more comfortable in using AI, way more than companies are and teams are within their operational practices. So they know they’re using their head is kind of what I would say first. So that’s not to say it should change our level of respect for privacy and regulatory standards or that we shouldn’t be transparent.
17:45
Meredith Johnson
But I don’t think we have to worry too much that people don’t know that AI is a really great technology that can help all of us in lots of facets of our life. So the candidates are ahead is what I would say. And then I think with, in terms of transparency, you might be in an interview and see a note taking. Right. So that’s transparent by nature of the interaction that you’re in. And I think that we, it’s fair to assume that applicants are using AI to support how they tell their story about themselves. They can assume that companies are also using advanced technologies to help understand who they are and to present their story and their brand in a really compelling way to applicants for specific jobs.
18:24
Meredith Johnson
So I feel like there’s some general knowledge there and I don’t worry so much because I think candidates, you know, are ahead of the curve. I do think, however, though, it’s important to note, you know, that you want as a company to also be transparent about your standards, your mission, your company policies, your brand, you know, in terms of what your beliefs are, how you think about privacy, what your culture is, how you treat individuals. Right? So I would say, you know, you should be bringing that into the interview process and actually it could be a key differentiator for your company. And I think being transparent, showing respect, giving a great candidate experience, and treating private information as private are standards that every company should be doing.
19:07
Matt Alder
A lot of the solutions that you talked about, you know, we’ve talked about how we can sort of implement them in workflows and all that sort of stuff. But, but each one in its own way is quite disruptive to the way that people think about recruiting and how recruiting has always been done. Things like, you know, as you said, note taking, that kind of stuff. I mean, how should TA leaders be managing that with recruiters and I suppose in particular hiring managers in terms of getting consensus around these changes and getting people to really buy in and adopt the technology.
19:40
Meredith Johnson
Yeah, you know what, I think it’s good disruption, first off. So when we’re talking to teams, and by the way, we have these conversations internally too, we are using new technologies and AI to improve the way that we work. Right. To improve the speed and quality of how we code, to improve the speed and quality of how we interact and track actions and meetings. Right. Like there’s all types of tools that we all use. And so we talk about this internally too. You know, I think it’s important to a couple things. One, talk about what again going back to those objectives, what are we trying to do, where are we trying to improve, to save money, to speed up our work, to free up time for more strategic thinking efforts and decisions and really be grounded in what are trying to do. Right.
20:23
Meredith Johnson
And then you have that conversation on well, what tools and what features are really going to help us here, right? What data, what tools, what features? And I think, you know, grounding in those objectives helps people get over some of that change engagement of using a new tool or using a new technology. And so I think that’s kind of step one. Step two is I actually think it’s fun to talk about how these technologies helped our help our teams adopt standardized processes and workflows. Right. So if you think about the power of structured hiring, which Greenhouse obviously deeply believes in some of these simple embedded right where you are features, particularly the AI ones, they’re not only beneficial for the efficiency and efficacy of the work as we’ve talked about, but they actually help drive adoption.
21:08
Meredith Johnson
So I’m sure we’ve all had challenges getting everyone on the recruiting team working in a similar flow, getting hiring managers to use the tool and to follow the structured hiring process. They want to engage and document important information in the scorecard. These AI features actually help with that because it makes it so simple. Right. As we talked about from setup to filtering candidates through the interview process, all the way through to offer all of these just moments of ease and support are really beneficial. But they actually help drive the adoption that sometimes has been blocking us to get that scale we want within our company, particularly around recruiting. So I see it as actually a really great opportunity.
21:47
Meredith Johnson
And so again, I would ground in the objectives we’re trying to achieve and then tell teams listen might be a little bit different, but here’s why it’s powerful. It’s going to help us in this way and it’s going to help us all stay in a flow and work similarly and follow that structured process that’s important to helping us get to that high quality candidate.
22:03
Matt Alder
And I wanted to finish off by asking a couple of questions about the future. So in the last couple of weeks we’ve seen again a plethora of new tools and new versions of LLMs and various other kind of AI technology launching. I don’t even know why I’VE said that because that’s, it’s an every week thing. It’s an every week thing. Now, is there anything sort of in particular amongst those emerging tools or things that you can see coming over the horizon that might really sort of impact the way talent acquisition works in say in a couple of years time?
22:32
Meredith Johnson
Yeah, absolutely. Well, first I want to say I hope everyone is starting to see and believe in exciting experience that AI is here and the value is actually already being realized. And so that’s really exciting. And I say that to say I think it’s proof for there’s more to come. And what is to come I think I’m mostly excited about. Right. I think there’s some really neat problems that we can solve together. And so when I think kind of at a feature level, I think some of the things that we’re going to continue to see show up is AI technologies that help with improvement, employer branding and career side experiences. Right. So really just setting the tone of who is the company and who, you know, why are we different, why are we special and why should people want to work here?
23:12
Meredith Johnson
We’re definitely going to see more capabilities in AI screening and interviewing. So for all types of companies, whether you are only hiring, you know, tens of people or hundreds of people and it’s, you know, very strategic knowledge worker versus super high volume. I think we’re going to see more prevalence of AI screening, more for chat and voice. So helping really upfront kind of get that volume started through the funnel. I think that’s going to be really important. You’re also going to see, I think interview automation and co pilots, especially in the high volume use cases, you’re going to see things for both candidates and our teams around interview preparation. I think that’s coming. And then really, you know, continuing to add value to that more personalized candidate experience. Right.
23:57
Meredith Johnson
So simplifying, you know, custom responses and pulling data out of the system automatically for you so that you can have that more tailored, meaningful candidate experience, which is really important. Right. And we’re in this competitive state once we kind of get down through that filter and it’s really important. And then I also think as we get toward decisioning and offers more support on debris and decisioning so we can feel confident in the decisions we’re making in those high quality hires. So really again, across that whole recruiting workflow, I think we’re going to continue to see new technologies, new features and solutions coming to market, which is really cool. I think during that time and beyond that, recruiting is going to get more and more agentic. Right.
24:40
Meredith Johnson
So we are going to see the prevalence of chat interfaces and assistance in our system, in our workflow. We’re going to be able to ask questions about data, we’re going to be able to ask summative questions about candidates or pools of candidates. And I think that’s really exciting. And I think, you know, if we’re here a year from now, Matt, or two years from now, I think we’re gonna be saying maybe we should schedule that. We’re gonna be seeing some really neat technologies that are really going to transform and to use your word, disrupt how we’ve always worked. But I think it’s going to be really great and really important. I think it’s gonna be great for companies and teams. I think it’s been great for candidates.
25:17
Matt Alder
No, 100%. I think I’m really looking forward to the way personalized candidate experience develops. I think that is such a big thing for the future. And I suppose just finally, I mean, where do you see the sort of the human AI split in all of this? What might TA teams look like in. It’s only 24 months time, isn’t it? But that’s still a lot of time for change.
25:38
Meredith Johnson
We need really great TA teams to make really great decisions. That’s not going to change. Right. I think what’s going to change is a lot of the burden that has kind of piled up over time, right. In terms of just the administration getting people to work in the same way to benefit from structured hiring. And then this cycle that we’re in. Right. This cycle is probably going to change, but the cycle we’re in, and I think we’re in it for a while, is that there is this extreme prevalence of applications and misrepresentation. Fraud is extreme. It is on the rise. Right. I think Gartner put a post out there that said in the not too distant future, almost 25% of applicants are going to be fraudulent. That is huge. That’s not going to work for our teams.
26:20
Meredith Johnson
We have to come up with solutions that bring this AI enhanced talent matching, that bring proactive AI alerts to misrepresentation, spam and fraud. And those are things that Greenhouse is focused on. Humans aren’t going away. We need to be on these teams. We have to make the strategic decisions, but we need help to get through all of that volume and we need help to defend threats that are just increasing. I mean that is the truth of what’s happening in the market. And so I think that relationship between human and technology, I hope as a technologist, it’s going to become more symbiotic, it’s going to be more natural, more embedded. We’re going to be soon over this hump of kind of, you know, concern or, you know, not quite sure how quickly I need to like, adopt these new technologies. Right.
27:05
Meredith Johnson
That will be a feeling we no longer have. And we’re in this more like high adoption culture of the technologies and we’re in a phase where we’re actually getting pretty nitpicky about what we’re going to use and what we’re not versus just should we use them at all. Right. So I think that’s really what’s going to shift is it’s going to be less about should we, shouldn’t we? And it’s going to be more about how are we doing it and how can we learn and share from one another and how can we get through whatever the next theme and challenges that’s going to face our sector.
27:34
Matt Alder
Meredith, thank you very much for talking to me.
27:36
Meredith Johnson
Oh, this was so great. Loved the questions and listen, this is so important and I really am very passionate about what TA teams do all around the world. And we want to diminish the impact of challenges and we want to make the work, you know, that’s done more. More meaningful and more high quality. And that’s what we’re here to do. So great questions. So glad to have been here and I really appreciate you having me.
27:58
Matt Alder
My thanks to Meredith. You can follow this podcast on Apple Podcasts, on Spotify or wherever you get your podcasts. You can search all the past episodes@recruitingfuture.com on that site. You can also subscribe to our weekly newsletter, Recruiting Future Feast, and get the inside track on everything that’s coming up on the show. Thanks very much for listening. I’ll be back next time and I hope you’ll join me.
