5 key takeaways for HR leaders

1. Leverage AI strategically

Focus on how AI can help achieve your organization’s strategic goals, not just operational efficiencies. Explore how AI can identify talent gaps and guide hiring strategies.

2. Manage change proactively

Address concerns about job security and help recruiters understand their evolving roles. Involve the team in the AI implementation journey.

3. Pursue quick wins with AI

Use AI to improve candidate experience through timely feedback and communication. This can build confidence and momentum for further AI adoption.

4. Ensure responsible AI implementation

Work closely with technology providers to understand and address algorithmic bias. Involve legal and compliance teams to ensure ethical and compliant use of AI.

5. Expand HR’s role in automation

Consider leading the automation efforts within the organization, leveraging HR’s unique position and understanding of workflows and talent. This can help drive strategic transformation.

HR.com 0:08
All right, welcome everyone. Thank you for joining “AI and talent acquisition: From automating tasks to strategic transformation,” sponsored by Eightfold AI. It is now my absolute pleasure to turn it over to Teresa Wykes to get us started.

Teresa Wykes 1:07
Brilliant. Welcome everybody. Thanks for joining us today. A quick introduction from me. My name is Teresa Wykes. I’m part of the Talent Center Transformation team at Eightfold AI. So, I’m delighted to have the opportunity to have two amazing people in my world join us today to talk about AI. It’s a very big topic. So you know, we’re going to focus on a few specific areas, but I think probably before we get started, Elke, would you like to introduce yourself and then Hung you can introduce yourself after?

Elke Manjet 1:46
Thanks. Teresa, excited to join today’s AI topics and technologies for the past few years, has made some experiences herself and is always keen to see experiences that other companies are making. I think it’s a big topic, as you said, Theresa, so it’s good we have enough opportunity to exchange about it. I’m leading the Talent Team at UiPath, a company that is in the midst of automation and AI, so from that perspective, glad to be here today.

Teresa Wykes 2:21
Thanks, Elke. Hung?

Hung Lee 2:24
Hi everyone. My name is Hung Li, so I still self-identify as a recruiter, even though it’s been a decade or so since I last sent the CV in anger. But these days, I work really more to consider myself as an ecosystem support person. So I see my role as really trying to connect everyone who’s involved in the act of recruiting, whether you’re a recruiter, HR person, candidate, technology company, events business, let’s get the conversations going between all of those folks and see whether the information transfer that takes place from that can just help everyone else get a little bit more inspired and a little bit more equipped for the future that we have ahead of us.

Teresa 3:08
Excellent. Thank you. Hung, so great representation from various sides of the coin. Here we’ve got Elke, who’s working in house, and then we’ve got Hung who is a kind of, I’d say, a great curator of all sorts of information. So this is a massive topic. I think, I think probably the best, and I don’t you know, we’re going to have a fairly informal conversation here. So this is, this is a chat, hopefully some of it be quite robust, not an interview. I think maybe a good place for us to start then is, is really, let’s start at, sort of at the beginning. So we’re here to talk about talent acquisition, but I think let’s maybe stay close to our purpose, which is the business context. I don’t know. Maybe we can start thinking about how we should consider AI in a business context first and foremost, and perhaps elk, maybe you could talk about what you’ve witnessed or what you’re experiencing from that angle at UiPath.

Elke Manjet 4:10
Yeah, I may talk about my time at SAP as well as we experimented with AI in ta there already. And I think if you’re already on the journey, you have started it, or you reflecting to get there. I think the starting question for me is, are you just trying to get some operational efficiencies, which is an important topic for all of us, or are you embedding what you’re trying to do with AI into achieving your strategic goals and the lather one goes beyond, let’s say, increasing speed of hiring, that could be a goal. But where I wonder is, are we getting to the point where AI will identify. Define the talent gaps for us, where AI will guide us to say, you know, for the skills that you’re looking for, this is the best place. This is the best market to tap into. So those are the more the whole skills topic is obviously a big one, right? Where I believe that in the future, TA and talent development will need to work much, much closer together if we’re all building on a skill foundation and maybe not focusing that much on the classical job descriptions any longer. So I think those are all bigger strategic questions in in the context of a company’s aim to get the best and maybe most diverse talent, and that is where I must admit I haven’t seen that much impact of AI yet. I think we’re at this point, still optimizing operational stuff, which is a good point to be. I don’t want to diminish that, because I think it gives us the room and the space to focus more on strategic tasks, and that is, from my perspective, the big opportunity that we have.

Teresa 6:15
Okay, do you want to talk a little bit more Elka about where you were seeing some of those processes, perhaps at SAP being refined by the introduction of AI, or even at UI.

Elke Manjet 6:25
So where, where I started to really experience this tangible. Tangibly was with a tool that we implemented to pre select applicants and gave the recruiter a ranked list of applications that meet best a certain job requirements, right? So that is the first time that I kind of really, I’ve seen it in practice that was back in 2020, I believe. So maybe some somewhat early, and I think some learnings from there are still true today when I’m talking to some of my peers as well. And I would say the first one is, don’t expect miracles. It’s not that you don’t plug it in, and the world has changed overnight. You may need to refine some things in your approach, in your process, in the underlying data. And I think it’s very wise to make your team part of the journey as you make your customers part of the journey, right? Because if you’re, if you’re, if you’re in a position that you have decided for a technology in this situation, we have the recruiter being required to give a thumbs up or thumbs down to the pre selection that the tool has done now, making sure recruiters understand that and understand that that five seconds investment is really crucial, because The AI needs to learn in the specific context. I don’t think you can buy one that does exactly what you wanted to do in the context of your business, of your company, or of your job requirements. So making sure that people understand it’s a journey there. The world won’t be better overnight. Everybody needs to contribute to a certain extent for a while, until it really yields the outcome that we want. And then one aspect that is important as well. Here from a business context perspective, if you’re not working for a company that is coding these AI tools themselves, you have a financial investment, right? So you’ve made a business case, somebody has approved that, and then you’d better be in a position six months later to say this is what we achieved. So from that perspective, I think the KPI aspect is important as well. I think it’s crucial to define what you want to achieve, and monitor that, and then be able to call it success or failure, maybe, which I think, especially in this context, still, still needs to be accepted as well. Okay,

Teresa 9:14
Excellent Hung. Did you have anything to add to that in terms of that business context? Maybe examples that you’ve seen.

Hung Lee 9:22
Ithink Elke has really covered it really well. I mean, that entire sort of monologue really is kind of encapsulating what the current challenges, what the opportunity is, what’s what I think is interesting for me is sort of the how, why is it that companies are really struggling to get there? And I think 2024 has really been that year where probably most people have accepted that there’s going to be significant opportunity with the use of artificial intelligence in terms of improving business efficiencies and operational processes, and even things like candidate experience and all. Those are more intangible stuff. We know it’s going to have these plus impacts there, but the vast majority of companies have not moved on from that, just that acceptance to operate operationalization. So what is the thing that’s stopping our product operationalization from going on? And I think Elka might have mentioned a little bit, it’s a classic change management type of, type of scenario. And, you know, we need to understand that asking people to change processes that they’ve been successful at and actually efficient at is going to be, you know, a painful process like there. There has to be some sort of acceptance that there’s a switching cost where your operational efficiency will decrease as you’re moving and learning into a new process. And I think that’s the crisis we have sort of in 2024. How that crisis sharpens in 2025 is I don’t think we’re going to be able to stay in that liminal state for another year. I think the business is going to expect to have dramatic improvements in terms of how recruitment operates. And we can’t have another year saying, Look, we haven’t got the bandwidth. We haven’t got this that the other we really need to move forward at that point. So trying to unpack why that is, and maybe, you know, trying to figure out certain ways in which we could advise people to block some of those obstacles. I think it could be a very useful thing for everyone to try and do. Yeah,

Elke Manjet 11:29
Can I build on that? Teresa, please go ahead, just in terms of of the change management, because I believe it’s, it’s very, it’s very fundamental to address as well the concerns that people have, and I feel like sometimes we’re just like getting getting on this ship with a lot of excitement and curiosity, and that’s good. But I think if you’re implementing these tools, people may feel you’re taking away part of their work, at least, right? And then I think this is a big task for leaders, primarily to make sure first, they acknowledge that that is the case, and we want it to be the case, right? Because we want those repetitive tasks to go away ideally and be done by technology at the same time, I believe we need to help recruiters to understand, what are they doing then, because the fear may actually be, I’m losing my job, right? I think we’re, by far, not there. And again, I think this is an opportunity for for recruiting, specifically to finally, potentially get into that talent advisor role, where we have all the data at our fingertips, where we have the process in the background taken care of by technology, and you can focus on hang touched upon it, candidate experience, really briefing candidates for the interview, good. Debrief. I think we’re all still hearing that applicants are disappointed to some extent with the experience that they’re getting, never getting feedback, being ghosted. All of that stuff is out there, right? And that is the opportunity for recruiters to tap in and change that, change that experience significantly?

Hung Lee 13:22
Yeah, AI can definitely help with that CV black hole experience. You know, I don’t want to prescribe the sort of things that the recruiters can do, but probably it’s now become a low hanging fruit for us to try and get a quick win. Because one of the biggest problems that we’ve had is obviously, you know, I think recruiters have have the good will to get back to people and, you know, do a good job, but it’s overwhelmed by different things to do, and inevitably goes sort of down in our priority list, and we end up making human errors, or it skips us, and we’re not able to get there. But artificial intelligence, like one of the base use cases, should be able to vary information to the candidates and to anybody in the process, without any latency, without any delay, without any error. So the AI can do this. And I just wonder whether, you know, that may be one of the quick wins. I see Angela’s mentioned something in the chat there. Where are the quick results from Ai? You know, maybe there’s a little project we can do there to try and, like, improve the candidate experience. Can we give them sort of feedback within 24 hours upon them performing this task or being in this sort of event, in the recruiting flow? Like, just focus on that and see whether we can move that number. And I think tech is generally there that can help us do that, and that will give people confidence, give the candidates a big sort of boost, and you then have some data to go back to your business and say, Look, we can make these improvements. We need more time to do this. The next thing.

Teresa 14:53
So, so, yeah, yeah, it’s interesting. What you were talking about, Elke with the whole change manager. I think there’s some sort of, there’s some general advice along the lines of, when you’re looking at your workflows, you should always invite AI to the table. You know, as in, you should, you should, you should consider it, bring it along. And then this, it sounds a bit of a cliche now, but you’re needing to be the human in the loop. You know, define, actually what that rule should be. And then I read something the other day that you should be treating AI like an intelligent but inexperienced intern. So, you know, it’s that it’s there to serve us. It’s not there to take over from us. It’s there to complement, and there’s obviously different degrees in terms of, you know, whether you’re using it for automation or it’s an agent, but, but hung, you know, in terms of where we are with it, I think, I feel that it’s, it’s going to get more, you know, I think the versions of AI that we’re using, like, if you’re starting now, is going to get a lot more sophisticated in the future. So it’s not like, like, whatever we’re doing now, there’s, there’s going to be, you know, we’re going to go much, much further down that, down that road. I think it’s also interesting to think about, actually, okay, we say that AI is going to free us up to do all these jobs that we couldn’t do before. Two things there, I think the first thing is that actually AI gives you data, or AI does jobs that no one did before, actually. So it’s not like we’re just, you know, we’re just, we’re just sort of passing off things that we don’t want to do anymore or that waste our time. It’s actually completely different data, completely different tasks. It does work that no one ever did. But secondly, where does that leave recruiters, actually? Because I think the trouble with a lot of recruiters now is that some of their roles are necessarily quite transactional. And I think being an advisor or being consultative, or maybe, you know, flexing emotional intelligence muscles, or really properly, deeply engaging with candidates is actually quite different from the role that they’ve done before. So, you know, I wonder whether actually, that would be something that we know we need, we need to think about.

Hung Lee 17:18
On that note, just to jump in, two years ago, I asked ChatGPT to rank the 100 skills and activities that a recruiter currently performs, and from there to strengthen one to 100 as to which ones that ChatGPT felt it would basically be able to replace. And it was actually very interesting to just observe what the AI thought it was capable of replacing, and what the AI thought it was not capable of replacing. And I took the top 20 of each category. So in other words, the top 20 skills and activities that AI thought they could basically do, and then the top 20 that the AI thought they couldn’t do, and just mapped them. And it is just very obvious that the work that recruiters currently do was exactly the areas that were most exposed to AI, in the sense that AI thought it could eventually do this. And there’s a bunch of stuff that actual recruiters were quite unfamiliar with, you know, I thought, actually, no, you know, I don’t feel particularly confident that I could, I could do this. So the job is going to transform. I would say it’s a good signal. I think the job is going to transform. We need to be active agents in transforming that, I would say. So in other words, we need to take positive action to change it. And I do believe there’s, there’s, there’s, there’s both opportunity and risk in all of this, because it’s, I think, quite naive to just assume that the time saved by artificial intelligence will automatically transmute into this higher level work that recruiters will just instantly end up doing. Because the business does not think like this. The business in 2024 has been asking TA to do more with less. Okay? So even in a company or a function that has done more with less, next year is going to do even more with even less, right? I’m going to keep on doing this. So what we do in TA and HR, and in fact, every department that’s exposed to AI, which everyone has to do is to negotiate back with the business what to do with the productivity gains that we’ve achieved. If I can go to my boss and say, Hey, I’ve just saved your hiring managers 5000 hours of time this year. That translates into this amount of money we need to have that investment back in the things that you say you care about, like diversity and inclusion, let’s say or improving candidate experience, or, let’s say improving employee engagement, and all of those things that very often are dismissed, or very often, you know, we the s that we rhetoric. Say their priorities, but when we run into adversity, the first things that drop off the list, we in TA and HR, have got to argue for that, and we can’t be naive to think that just being efficient is going to lead to us doing that work. We have to basically have some consciousness about what that extra time is going to be, and we have to negotiate for it.

Elke Manjet 20:23
Yeah, I would. I would totally agree that young you need to have your business case, and you need to tie to your strategy, and I think that’s where you’re getting the buy-in from the business. And obviously, as you said, showing them where you can improve when you can, where you can save time, not only for TA, but as well, for the business leaders, who sometimes are quite frustrated with how ta processes work, right, how much investment is required from from their side, even to the point, you know, of writing job descriptions, which really shouldn’t be required these days any longer, but there is a few of those very basic tasks that are still happening sometimes on the business side. And I must say, I feel wisdom, and I think again, that’s an angle for us to do a good job in terms of making sure we have everybody in the change journey to understand their contribution. But as well as their benefits.

Teresa 21:22
Got a couple of comments and questions, so we’ve got a question here. Rory is interested to know whether we’ve ever encountered any algorithmic bias with AI screening, and if so, what did you do to counter this? Elka, maybe. Do you have anything to share on that?

Elke Manjet 21:41
So for that piece of technology that I talked about in terms of pre-selecting applicants, the KPI that we set for ourselves is that ultimately, 75% of the applicants that the technology rates as a B, like the best ones, need to end up in interviews. We defined that as the quality criteria for the technology, and we have seen for a few weeks that that was not the case. Like we had applicants ranked A, B, didn’t, weren’t selected by the hiring manager to come into the interview, so obviously they the AI did something wrong. So we found out that in terms of how we describe criteria in job descriptions or the job requirements, that was not exactly helpful to how the AI is working. And we actually with the company at that point in time that we worked with worked closely to change that they ultimately did adjust the algorithm to some extent, and we adjusted our job requirements definition, and from there, we’ve seen a different outcome, but it took us a while to get to the root cause of the problem, which, again, brings me back to my point of, don’t expect miracles. You really need to have the thing in practice, in your context, to see what it does, in terms of what you want it to do, and where there may be a gap that needs to be closed.

Teresa 23:18
Yeah, and I think that’s the thing, isn’t it, you can actually tune the AI to your requirement. You know, you do. You do have that agency. You’re not, you’re not like you know, we’re not slaves to it. It works for us, it serves us, that you can, you can, you can change things.

Elke Manjet 23:32
Yeah, that’s true, Teresa, but I would want to add maybe one comment to that, because I think, you know, we’re talking about how our work is changing, and I think in that context, where we all need to be prepared to go deeper, is on understanding what the technology does right. Because there may be nobody in your company who can help you. You really need to work deeply and closely with that provider to get to the root cause and get them to fix it. So I think our our whole kind of stakeholder map, maybe maybe a bit more intense in this context than in the past, as well, for these type of implementations, you would want to work closely, obviously, with it, but as well with your legal function, for example, because, let’s say it does discriminate worst case. You know, you need to be prepared for that.

Teresa 24:26
It’s interesting, isn’t it, because it forces you into the world of technology in a way that you probably thought you wouldn’t need to bother with before. How Helga? How did you maybe, maybe, when you were at SAP, how did you decide how far you needed to go with your own personal knowledge? Because you need a level of understanding, don’t you, but you don’t need to become an AI specialist entirely. I guess you need, it’s the level of awareness, and also interested to know what kind of support you had around that from the rest of the business.

Elke Manjet 24:59
Yeah. Yeah, yeah, I think you’re touching a good point. Teresa. You need to pull in your network, right? I’m likely not going to be an engineer in my professional life any longer, I guess. So there is a level where you can push for understanding, ask the right questions, go into depth, go into detail, but you get to a point where it’s just beyond your knowledge and your understanding, right? And I think that’s where it’s key to to involve your partners. So to say that is it, if you have an AI engineering team, why don’t you invite an engineer from that team, because they know best, right? And you may have, you may have a data protection officer or some, some type of that function that that’s a good partner to involve, and then again, compliance and legal, because this quickly gets you into an area where you need those specialists to make sure what you’re doing is not creating any issues down the line, yeah, as well. From a candidate perspective, right? From an applicant perspective, what happens with my data if AI on a company side is working with my data? It’s a big question, right? So you very operationally may need to update your communication to applicants on we have AI in the loop, and this is how we safely work with your data.

Teresa 26:29
Yeah, yeah, yeah. I think Yeah. There was a Harvard Business Review Report that, and then hung I want to ask you about something, because I know you have a view on this kind of role of chief automation officer, as well as a sort of new role in the world of in the world of AI Harvard Business Review said that leaders should be able to provide context, identify content, lead with their heart, and ask good questions. So in other words, you know, it’s not so important to discern all the answers or even to know how to leverage the algorithm. So I think, I think that’s probably something that’s going to, you know, develop and will always be on leaders minds. But hang on, you know, you talked a bit in some of your recent newsletters about having this, you know, organizations might need a chief automation officer. Do you want to? Do you want to talk a little bit about that, and where you see that maybe meeting needs and maybe not meeting needs? Yeah,

Hung Lee 27:32
Absolutely. I mean, basically, one thing we do know is that the sea level is very disappointed at the rate of automation or the use of AI at every level of their business. So there’s no bigger enthusiast for artificial intelligence and automation than sea level. They can see the productivity gains, and they’re feeling really quite disappointed. The heads of function are not delivering for the reasons that we’ve discussed. There’s a change management issue. This is basically hot swapping one system to another system. It’s, you know, we’re already massively under pressure, very difficult, so we need to unpack that. But I think there is a case to be made that there may be a specialist function that might emerge that will actually go and start examining sort of people’s functions, people’s workflows, and see whether automation can particularly improve that, or whether you particularly want automation to improve it at that point, it could be, for instance, that you decide that you don’t want to have a highly efficient, overly efficient way of working, because you might lean a bit more on, let’s say, improving experience. So automation may be required, so the failure of all of our functions to automate may mean that a new kind of function might evolve that will help other departments do it. In other words, conduct the analysis as to how you know, time and motion studies, how the thing is currently done. Now have understanding or with the vendor landscape or the technology sort of innovation landscape, to then be able to recommend solutions and maybe even project manage them in ways that is a classic change management type of function. So chief automation officers, and my idea is an idea friend of mine actually published what two years ago and said, yeah, maybe this is an idea. Thought was a really good one. And then I thought, actually a lot of the responsibilities within that hypothetical role kind of overlap a little bit with a lot of the stuff that people in the people team do you know, understanding how people work, understanding what the job specs are, and so on. So I just wonder whether that may be a kind of a way in which TA or HR might expand scope to become an automation force within that organization. Because if we don’t do it, basically, they’re going to call the consultants in and they’ll do it. You know, if you look at all of the big four big consultancies, the McKinsey’s, the Accenture, Sue Wise, all of that, all of those companies, are not actually performing particularly well in terms of their margins and. Apart from the AI functions within each of those businesses. So there’s a huge demand for companies that are going to these consultancies and saying, help us automate. Help us get more efficient. Now, no disrespect to that function, but as soon as they walk in, they’re going to start making changes and saying, You’re not automating fast enough, and then we’re going to be completely at the mercy of those instructions. I would much prefer that automation functions organically emerge within the existing business. I see no reason why someone in TA or HR can’t do that.

Teresa Wykes 30:34
Interesting.

Elke Manjet 30:36
If I can add to that Teresa, I think there’s again, sinking in opportunities rather than risks. There is a unique spot that HR is in, in the sense that many of our tools are touched by everybody in the company, right? A ta tool is not touched only by TA. It’s such by internal employees who apply. It’s touched by hiring managers, interviewers. So I think that that is an opportunity for us as well, and getting a bit of a call for courage and risk taking out here to kind of lead the change to some extent, right? Because if we bring it, it quickly touches everybody, and that’s a great opportunity to show how we can kind of step ahead potentially.

Teresa 31:28
Yeah, yeah. Thank you. Just checking your questions again. We’ve got a question from Angela here. She says we have very basic AI tools like ChatGPT right now for our recruiting team. And we’d like to see in what ways you can use that to make more efficiencies in TA. So, you know, chat GPT available to everybody. Anyone want to talk about how that can make a difference?

Elke Manjet 31:57
So I can start and hang maybe, maybe you can, you can add, I think we’re all using it to some extent, for us individually, right? And obviously, if I’m in the whole communication with applicants, hiring managers and so on, it can help. But my view is, you know, make sure it’s personalized enough for that situation and for that context. So what I think would be the best approach is to personalize it for your applications, with your context, with what you want to bring across as a company to the talent you’re trying to attract. And for that, it probably needs to be a bit more specific than what ChatGPT gives you. Is this my experience? Not sure of any, any views on your side? Yeah.

Hung Lee 32:55
It’s very interesting how we kind of encounter AI sort of as professionals and chess, ChatGPT was probably everyone’s first, AI sort of say, and it was a miracle solution. It still is. It’s fantastic for things like copywriting and, you know, content generation and even ingesting large amounts of text and having that, you know, summarized stuff and all the rest of it. I think what is interesting to me is that all of the key features or functions of artificial intelligence are starting to become very easily available in multiple places. Chat, GBT initially was great at message composition, let’s say, and then suddenly LinkedIn can do that, and Google’s now being able to do it, and Microsoft can do it. So what we’re seeing really is a lot of the generative AI basics, so to say, are now being encountered in multiple places, not just chat, GBT. And I think what we’re likely to do as recruiters is that we’re going to kind of use the predominant software that we’re always kind of either needing to use due to compliance reasons or simple cultural habits, and that product will end up having huge amounts of AI in it, and that’s going to be our default interface. So let’s say using an ATS every ATS right now has just lead pumping out AI powered features at you. So you’re going to be AI enabled, even if you don’t, sort of, even if you’re a refusenik, you’re going to end up using these types of tools. Now I don’t think this is a good or bad thing. I think where we’re at with us as professionals is that we just need to be very, very fluent with the usage of it. We need to stop, like dipping our toes in, but really spend time to get the maximum juice out of that lemon. You know, the worst thing we can do, I think, is, and I’m guilty as everyone. By the way, I’m not some expert user of AI. I find myself struggling to, you know, because I’ve got to exist. Non AI system, and I’ll defer to that, but I know that that is an inefficient way, and I’ll be out-competed, if not tomorrow, then we’ll be in May, or will be in three months, or six months, or some other months down the track. So I don’t think I’ve actually answered the question here, really, but ultimately, what I’m saying is that the leadership of TA teams have got to create the bandwidth for the TA teams to feel that they can, they can experiment. And this is very, very difficult to do, because we’re all working at 120% maxed out capacity, right? So what we’ve we’re kind of driving the car 100 miles an hour down the freeway, and we’ve got to swap out all the tires. That’s what we’ve got to try and do. And we need a pit stop, really, to do that. But in order for us to do a pit stop, we have to talk to the business and negotiate, as I mentioned to you guys earlier, like if we can negotiate a deal with them to say, look, we’re going to change that process. There’s going to be a temporary injection of inefficiency as a result of that. But what we expect to have happened once we’ve done this is that these things will occur, and if they do occur, then I want more backing for us to do more of this kind of shopping. And that’s the sort of conversation that leaders, I think, need to take on board.

Teresa 36:11
Yeah, 100% 100% and I think, I think that we’re going to see a bit of a resurgence of some of these sorts of more maybe some skills that we’ve kind of forgotten, like problem solving, just learning, you know, just actually having to learn stuff. I think there’s a big gap in actually understanding the limits of AI, you know, because it’s not, I think, I think, you know, there’s, there’s, there’s a sort of misconception that it can solve all these problems, but it only, it only learns on what it’s, what it’s been fed. So, you know, there’s a sort of the ethical side of it. There’s also the hygiene side of it. And I think there’s a responsibility definitely for leaders, just to make sure that people actually appreciate what that, what that, what those limits and risks can be.

Elke Manjet 36:54
The question on the re-skilling piece, I think Hung has touched on that, and it’s a really important one I believe, and maybe I’m stating the obvious. But there is a sentence out there that I like which says, you won’t be replaced by AI, but you may be replaced by people who know how to deal with AI, right? So in that context, and as well, considering the constant evolution of the technology, that’s, that’s, that’s changing rapidly. I think we need to be prepared for constant rescaling, right? And probably people know, and I’m stating the obvious here, but there are opportunities out there in the context of citizen developer training, right? Microsoft is doing that. UI pass is doing that. Many big companies are doing it, and it’s an easy, somewhat easy way to get started. And what we’ve done at uni pass is really making sure that many of our colleagues have the opportunity to get that certification. And then we had once per quarter a day where people did come in after they had kind of the training, and they came with their ideas, and they have that day free to do their own little automation. And it was, it’s been a lot of fun for the team. It’s been a great exchange. And there was a real outcome in terms of processes improved and time saved, so that that could be just one approach that may work for people.

Teresa 38:32
That sounds great, that sounds really interesting. Please, keep the questions coming. I particularly just want to ask to talk a bit about candidate experiences here. So we touched on it a little bit a little bit earlier. But I think this, I think all a lot of roads in AI, if you’re in to lead back to the candidate experience, because I think it affords us a very different way of treating talent, but also how more, how much more empowered talent is when they are applying for jobs, making decisions about jobs, having data about things they never had data before. It’s turning work on its head very much. But I’m just interested if we can, if we can talk a bit about, particularly this personalization of candidate experience, while you’re also balancing your internal and your external hiring strategies.

Elke Manjet 39:34
Yeah, so I think if I may start, what AI can help us do more. And I know some companies are leveraging it already. Others may be more at a starting point that is really hyper personalized, the communication and the interaction with candidates, right? Meaning that if I. Applied to five jobs in a company. The AI, too would know that, and would address me in that way where I’m not just getting turned down for one job and then I don’t hear about the other four, and I wonder, does it mean I’m turned out for everything or right? So, so those are the things where, there’s still some flaws today, and where I think the technology could could help us in addressing people in a very, very specific and personalized way, considering all the data points that we have, that we have around them, similarly, if you’re if you’re thinking about talent nurturing, right, I think all companies experience hiring freeze hiring. Go hiring freeze hiring. Go right, that’s what’s tough for the TA team to manage. So if you’re going to do an approach where you say, we’re constantly nurturing talent, independent of RECs that are open or not, again. I think that’s a field where AI can help tremendously, because it’s always a challenge to keep track of who has been talking to a certain person, especially if it goes over a longer while, let’s say six months a year and so on, keeping the content of that conversation so if somebody else picks it up, you know, where has it stopped? What’s relevant to that person? Why have we engaged? Where are they in their openness to continue the engagement? So, that’s a very early part of obviously attracting talent, where I haven’t seen that much myself yet. In terms of technology helping there, but I could imagine that coming rather soon.

Hung Lee 41:47
Yeah, I’ve seen some early examples where that is exactly happening, the kind of activation of pre acquired candidates. So to say, I think this will actually become an increasingly big part of what AI can do, because the other side of it is that when you have to acquire candidates, it now comes up with, you know, a significant degree of risk. Anybody who’s posted a job in the last 12 months will tell you they get inundated with applications now, because, of course, candidates are also using AI, and they’re also personalizing, and they’re also doing mass supply. So I think pretty soon that we’re going to end up having a deterrent towards advertising jobs publicly, because we’re kind of inviting more work than we can tolerate. And so we’ll start looking at other ways in which we can find candidates. Guess what? You’ve already pre acquired candidates for every company. There’s 150 people listening to this. Guarantee you, if you added up all of the database records in 150 companies. So for people watching this, you get into the 10s of millions of records. We have accumulated huge amounts of data that’s just sitting there. Now what AI can do is reactivate that, that data. And this is classic talent relationship management, which, you know, pre AI, we had something like that, you know, with a little bit more sophisticated decision trees, type stuff and software that helped, you know, if this person did that, then we’re going to send this. But now, genuinely, we might have AI that’s able to have a much more fluid interaction, interactive conversation, quote, unquote, with every single record in there that wants to have a conversation with us. And I think that’s going to be a kind of a very rich source of candidates in the very near future, with very little human effort, to be honest. So pre acquired candidates, alumni networks, existing employees within companies, you know, in terms of internal mobility, talent, liquidity and that type of stuff. AI should unlock a lot of that stuff, a lot of that activity and so, so we as TA are part of the reason, again, we need to, you know, have the imagination to expand scope beyond just acquisition of candidates. Talent acquisition is externally hiring candidates in but the business doesn’t really care about that. It cares about having the human capital, the capacity to do a certain function at the certain time when it’s needed, and what we’ve got to do is be the function that’s able to supply that need. And so it’s much more about business capacity, supply, that’s what we’ve got to do. And acquisition, in terms of hiring external people, is one of those ways we can do it, but there’s now many other ways, and again, we have to embrace those opportunities, expand scope in terms of what we have to do.

Elke Manjet 44:39
Yeah, I would totally agree. And I think what happens there as well, is that, in the meantime, we’re seeing chat bots where we can’t, like differentiate, or we can’t realize it’s a chat bot, right? It’s so good that you’re actually feeling like you’re talking to a human being. For sure, there is room for advancement there. But I think. Seeing things, getting there, and that makes the experience for people a better one, a more positive one, versus a clunky one, where you’re desperately looking for that human being to interact with.

Teresa 45:11
And I think as well Elke and Hung that whilst at the same time you’ve got all of this, you know, all these, there’s much of this greater efficiency with AI in terms of your some of your talent pipelining, or your talent engagement work, I think at the same time, it’s going to create corners of need for real kind of white glove treatment with talent so, you know, talked about talent concierges or talent butlers, you know, people who who actually really wait. That allows you to be very selective about where you place, that very, you know, white glove treatment. And also, I think we’re going to see a nice return to recruiters as career counselors, you know, who are having proper advisory conversations, you know, pulling from that sort of more exec search type of approach, where you’re, you know, you’re giving advice to people, as opposed to sort of transacting your way through, you know, a list of questions. And I think that this is a fabulous opportunity to really define the role of the recruiter now. And I think it will professionalize it, and I think it will it, will it, will it, will regulate it, and probably lend a lot more, a lot more compliance around it as well. And you know, most people are very well informed these days. I think your, you know, your career is sort of in your hands, particularly as you as the data becomes a lot more transparent. So yeah, I think, and I think as well, on the back of that, we look at our jobs differently, like we’re now entitled to talk a lot more about fulfillment in our careers, in a way that perhaps, you know we didn’t so much before, you know, because we can make choices. So yeah, there’s a there’s a the world of work is changing way beyond you know, the activities that you know we’re doing in talent acquisition. Okay, so, so what, where do you see this going then, I mean, do you think that we’re kind of, you know, we talked a bit about the fact that things are going to get better and worse, depending on which way you, which way you look at it. Do you think perhaps Elka is there? There are parts of, parts of the world of talent, attraction, that that, that that could be, that could be even better served by AI, that that maybe hasn’t been thought of yet. You know, do you see what? Do you see that that world will grow and grow and grow? Or do you think that we’ve kind of seen as far as it needs to go right now?

Elke Manjet 47:44
I combined this Theresa with Tina’s question that I saw in the chat, where she’s saying we talked about pre-selection quite a bit. Can AI solve other pain points in later parts of the process? And if I can start with that, maybe, definitely. I mean, you have, you have tools sending out correct material for interviews, to applicants and to hiring managers, doing the follow up. I guess many, many da teams are still challenged with ensuring they get timely feedback after interviews from hiring managers. That’s something that AI can do for you. Some companies go as far as having AI doing the pre interview. That’s happening already as well, right? If you, if you go down the road, you can have a bot do your offer letter creation. That technology is, is there as well already, right? Obviously, you can have bots do candidate feedback and drop that, that is, those few feedback questions out to applicants at each step in the process, right? So I think there’s, there’s a lot out there already, and with technology progressing, I think the opportunities are probably infinite in terms of the sheer technological capacity, but I think that’s the space for us to shape things and to reflect like, where do we want the Technology to take over, versus where do we want that human in the loop, or maybe more than just in the loop, and still, you know, valuing that, that human relationship and that human interaction and that part of the experience versus the efficiency, right? I think hang touched on, touched on this earlier. So technologically, as far as I understand it, I don’t think there’s really a limit. You have onboarding that’s facilitated with AI, in the meantime, in virtual rooms where AI technology makes that experience much more exciting and real. So that’s out there already as. Well. So again, I think technology is infinite. It’s on us to really, to really define where we want to, where we want to leverage it, and what that part of the job is that we feel we still owe it to our customers. Be the applicants, be it, be the hiring managers, to still do it ourselves and bring that human touch in and that creativity and to some extent, as well, a bit of that assessment, if I may say, because I haven’t seen yet, AI that is uncovering the hidden gems, for example, so that that talent that may fit greatly to your role, but is not ticking the box on the classical job requirements, but it’s bringing some other skills, competencies, background that may approach the job from a different angle. That’s, I think, still a piece where I haven’t seen a technology that can do that, not saying it may not come, but that’s where we we still have a unique kind of value proposition as humans, I would say would be great, wouldn’t it?

Teresa 51:11
Yeah, interesting. Anything you would add to that home?

Hung Lee 51:16
Um, no, I think sort of Elke definitely covered, um, sort of the scope of the potential of artificial intelligence. I think people listening to this will think, you know where what’s left for humans. So to say, I’ve got a theory on it, and it’s kind of aligned with the last thing that elke mentioned. So firstly, artificial intelligence is certainly going to out compete human beings. Whenever there’s explicit information that’s structured and documented, it’s just literal base cases. So anything that we’re doing where it’s documented information, where it’s structured and it’s explicit meaning, it’s not inferred, it’s literally written down, then AI is going to take that on now, where human beings come into play, to where the information that’s implicit and undocumented, and that is the reading in between the lines. That is, you know, what do you really mean by that comment? What is the off-the-record conversation you have with the hiring manager? Tell me, really, why is it? Why is this job open and it’s off the record? All right, those kinds of conversations are often the most important ones that you have in your professional life, because they’re the deep motivations that have, you know, created those circumstances. And I think that we probably will never have AI dealing with that, we probably don’t want it, because that type of human interaction, human exchange, is what intuition is. It’s basically what institutional knowledge might be. If you look at, for instance, the failure of companies to immediately replicate the success of another company, even though they may have stolen the manuals of how to do something, or even with that original company literally saying, Do it like this? Other companies trying to replicate it don’t work. Why? Because there’s like, loads of holes that have not been documented. Now here’s the thing, AI is really good at everything that’s documented information, but human beings are really good at the undocumented side of it, which is, by the way, part of the reason why I’ve been traveling as much as I have, because I’m interacting with people in real life, in person. Why? Because I need to get that implicit information. I need to have that sort of peer to peer information that isn’t written down. I spend way too much of my life locked into the matrix, looking at a computer screen, imbibing information. That’s what recruiting brain food is. I dare say a lot of recruiters are currently doing this, but you have to complement and corroborate that information with in person interaction. And I think if we kind of commit to that, we’re going to end up with an edge. We’re going to end up as an AI enabled recruiter that has the edge because we also have access to collective intelligence, because we’re speaking to our communities, we’re speaking to our candidates, we’re speaking to our peers. That collective intelligence, plus artificial intelligence, I think, is the future for recruiting.

Teresa 54:19
That’s a very, very important thought to end on. So thanks for that hung. I think we’re just about at time. So remains for me to say. Thank you everyone for joining. Thank you to Elka, thank you for hanging some interesting discussions. Hopefully it was thought provoking and useful for you, and I’ll hand it back to Rhonda. Thanks.

HR.com 54:42
Rhonda, absolutely. And thank you. I definitely want to take the time to thank all of the presenters and all of you for joining us and to Eightfold AI for sponsoring the webcast. Just take a moment to fill out the feedback survey. Your feedback is super important to us and the folks at Eightfold and the presenters here today, and if you have come late or you’d like to view this again, it has been recorded just now, and it will be archived shortly on hr.com. The HRCI and the SHRM credits, those will be in your account within 24 hours, and we will also send an email to the live attendees with your credit information, and then once you fill out that feedback survey to go to the next session, you would just head back to the webcast auditorium and join the next one again. Thank you so much. Theresa Elke and Hung take great care everyone.