How AI Is Shaping The Future Of Skills
In the latest HR Leaders panel, experts discussed AI's transformative impact on HR, highlighting a shift from job-centric models to skills-based approaches. They explored AI's role in workforce management, strategic decision-making, and identifying knowledge gaps while addressing ethical considerations. Key learnings included leveraging AI for talent gaps, real-time intelligence, ethical navigation, and innovation acceleration.
Discover the transformative impact of AI on HR practices. Listen to industry leaders discuss the shift from job-centric models to dynamic, skills-based approaches essential for today’s agile business environment.
Top experts in HR, Talent Management, and Learning & Development shared how AI tools are revolutionizing workforce management by providing visibility into skills, facilitating strategic decisions, and identifying knowledge gaps.
The panel will also cover the ethical considerations and limitations of AI, offering a balanced perspective on its role in HR.
🎓 What you will learn:
Identifying and bridging talent gaps with AI
How to leverage AI for real-time talent intelligence
Navigating ethical considerations of AI in talent management
Accelerating innovation through AI-enabled talent reallocation
The world’s AI copilot for HR, powered by HR Leaders.
Say goodbye to uncertainty and hello to instant, expert-backed answers with atlas copilot. Our cutting-edge platform revolutionizes how you learn and access HR insights, providing instant answers to your HR questions leveraging unmatched expert resources for enhanced productivity, informed decision-making, and skill development.
Powered by HR Leaders, atlas copilot combines the knowledge from the world’s leading thinkers including Dave Ulrich, Amy Edmonson and Marshall Goldsmith with over 10,000 hours of video interviews with leading organizations such as Apple, LEGO and L’Oreal, and 50,000+ pages of research tools, articles, frameworks to offer the most-comprehensive knowledge bank for HR professionals and executives.
Chris Rainey 0:06
Hey valon. Good morning. Good afternoon. Good evening, depending where you're tuning in from. My name is Chris Rainey, co founder here at HR leaders. And I'll be your host for today's live power discussion, where we're going to be talking about how AI is shaping the future of skills based talent management. With that being said, let me introduce you to our amazing panelists. First up, we've got David doe, who's the vice president of talent, strategy and excellence at Shell. We've got Sarah and Atolla, who's the global head of talent, culture and inclusion at racket. Mike, Michelle was a Coates, co founder and CEO of 365 talents, Alexandra Foster, who's a vice president of people experience at Kelly. And last but certainly not least, Tony Wiseman, who's a digital HR director at Columbia, what I needed to take a bigger, bigger breath there, run out of oxygen, as well, but, Alex, I'll pick on you first. Firstly, how are you? Let's say that first,
Alexandra Foster 1:01
I'm doing well. It's a good Thursday here.
Chris Rainey 1:03
Nice, nice, tough questions to jump in. But let's just jump straight into the deep end. How is the role of HR evolved in your organization with the shift of, you know, from traditional job centric models to AI powered skills based approaches?
Alexandra Foster 1:19
Yeah, no, it's a wonderful question. And I will say, first of all, I'm honored to be here with everyone. So thanks for having me. The wonderful thing about working in an organization like Kelly, where we're a talent solutions provider, is that you get to serve as both a testing ground right for new and innovative approaches. But you also have the benefit of hearing from a cross section of employers about what they're looking for what they're struggling with, as it relates to a host of talent solutions, but AI be one of those. And naturally, we've been on our entire journey for about six years. So naturally, when we started, our analytics and technology teams, they jumped right in, they were already building their skills, they knew what was coming, and had already been doing the work to think about approaching their roles differently. But as we talked about embedding more solutions for our people, we have some other teams come to the forefront. And one thing that I'm really proud of is our learning and development teams, I think part of it was that they wanted to be prepared for all the skills that they were going to help build across the organization in support of the shift to more AI solutions. But another part of it was they were eager to tap into all of the benefits of using AI more in their roles. They became sort of city citizen developers, they were raising their hands to test things like co pilot. And really, their roles begin to shift earlier than the balance of the organization. And what I think that's taught me about how HR role HR roles are going to shift in general is that we have to help the organization by moving into more of what I would call a fourth position being pilots and understanding being curious being chief, the chief learning officers kind of like see my role is as not helping others learn, but being curious, and learning about the new ways of working, learning about the technology, pushing myself helping to build skills amongst my team, so that we are experiencing this change together with our teams, and that we're not doing it to them. And so I think we'll continue to see more of that, and HR teams and HR functions where we are working side by side and navigating this change with our teams, but also at the same time. And we're looking for what's next and always looking around the corner. So that we are helping to maintain, you know, the healthy culture and address the concerns that these concerns that our employees have with AI. As a
Tony Wiseman 3:51
quick question there, that was wonderful. In terms of the organization and how you go, because the shift AI brings, obviously a lot a lot of promise was there's a piece word digital literacy or the digital fluency of your people within the organization kind of needs to be brought up to a certain level as well. How have we looked at our RV attack that that that issue where, you know, different levels in the organization would have different levels of fluency or digital fluency.
Alexandra Foster 4:28
I and I do believe that is what was motivating our learning team. So we have a cross functional team that is thinking specifically about AI and digital adoption at our organization. Our learning team is well represented there. But we've also launched recently, what we call alerting steer CO is probably a common practice at some larger organizations, but it's newer approach for Kelly because learning is decentralized. We've have some learning that is old centrally with our learning team and some that happens that is more kind of role based, and the shifts and changes that are happening more rapidly. So that work is deployed closer to the teams and the businesses that they support. But the steer co brings together a cross section of folks, and we're trying to anticipate what those gaps are much like everyone is, you know, I also lead the teams that are responsible for our employee listening strategy. And so this quarter around on our employee engagement survey, we have a sentiment question, we don't need leaders to activate on it. But that that kind of steering committee and AI and digital teams are looking at this insight, we asked employees, how open are you to using AI every day at work? Do you feel you have the tools that you need? And of course, we'll put that in, in our AI tool to get help us sense check. Where do we have opportunities to accelerate work? Because our employees are open? And they understand? Where are they feeling like they don't have what they need? Or that they need a better understanding of tools available, how we plan to use them? Where are the concerns about ethics. And so I think always opening up the channels of communication for Kelly is really critical and listening directly to our leaders, but also working with our employees, but also connecting with leaders, HR partners, our recruiting team to have an understanding of what what skills are in demand at any given time. But in this challenge, for sure. There's not 111 There's no silver bullets here.
Loïc Michel 6:24
Thank you. Thank you. That was great.
May I ask a question as well? To you, Alex? Sorry, Chris, you're not asking the question. Today?
Chris Rainey 6:33
Listen, this is great. I'm like Alex, I'm sorry. Like, Alex, I didn't know I was gonna get all these questions, by the way.
Loïc Michel 6:41
But Alex, you share that, as a talent solution provider, actually, you are, you know, in the front seat in the front row of all these transformations, and you are experimenting and testing and you are really mature. How would you describe the maturity of your customers in this, you know, in these transformations? Like if you had to give a grade from one to 10? And is it verticalized? Do you have some particular industry where you see it's going easier? And you speak the same language and other industries where there is a gap that is harder to feel like curious about this? Yeah,
Alexandra Foster 7:18
no? It's a great question. So it's been on an AI journey for about six years, maybe almost seven years, with the most mature vertical being our business to serve service delivery model, I would say our own internal HR model is maybe a lakh or two behind that simply based on our size and our volume, and where we're making investments just like every other, every other organization. And then we've also really recently released some products, so newer, but more mature, because our clients are pushing us there, where we actually have digital workers who accompany our talent on their assignments with our clients. And so I see, I see that there's a range, obviously, where you have industries that are making a shift in the tech space. In it, sometimes even in manufacturing, where they're looking at ways to reduce cost and to shift that work that's necessary and critical, but maybe arduous, repetitive, and freeing up workers to do more of that thought work and the creative work that that maybe only humans can do. We're seeing that product, vertical, accelerate pretty rapidly. So that would suggest that our clients in some spaces are really ahead of the game. But I'm also seeing that right and hearing from colleagues in a variety of settings in HR and technology and shared services that even small organizations are looking to optimize, I think, you know, sometimes we talk about leveling changes in the industry in the future of work. And AI is one of those great levelers because whether you're a large organization, you're looking to deploy deploy AI, or you're a small organization will need to capitalize some of the benefits, you're able to tap in and play. Of course, you know, we've got to have good guardrails, good guidelines, but it is a space where an organization that has the mindset and the cultural really readiness to use more AI, they can get involved, regardless of their size. So it's very interesting. What we're seeing is a great question. Thanks.
Chris Rainey 9:21
Yeah. I'd love to jump over to you, David, because I know you're already pretty not far ahead. But I say that for many companies that I speak to, could you share some of the ways that your organization's using AI to identify critical skills gaps, for example, and some of the steps that you're taking based on those lessons that you've learned?
David Doe 9:41
Thanks, Chris. I'm not certain I would necessarily characterize as right at the front of the queue. But, but certainly, we are in the process of piloting both a talent intelligence platform and XP and we've taken our time to think about how we would do that spoken across a number of peer organizations, we made a very conscious choice to take the talent intelligence platform, the taxonomy that comes with it and the AI, and use that as the sort of base layer that goes across both the HCM and also the learning experience platform. So that provides a unified taxonomy that allows us to go across these different, you know, programs, the challenge we found with other talking to peer companies is basically, each of these solutions has their own taxonomies. And you create a poor user experience. So we chose the talent intelligence platform, because it has that labor market connection, and that it moves close with the labor market. And of course, you have to go through a normalization process where you take your positions, your job profiles, and you feed it into the AI engine. And you find that, you know, you've got some interesting acronyms that make life difficult for the AI engine. I mean, it's like a child learning to read, except it's a child learning to read on steroids. So once you are able to tell it, where the what the acronyms are, it will quickly learn and adapt. But it's also important to them have good job descriptions, etc, etc. And we've also had to be really clear, what's a competence and requires an assessment. And in other words, the AI shouldn't be touching that. And we're actually it's a skill and we're prepared to let the AI do inference. And so essentially, the AI will match skills to people to jobs, and to learning content. And it can infer when someone ingests their work profile, essentially the skills for that individual based on their experience, the breadth of that experience in a number of roles and the recency of that experience. And so that's an initial inference, and the individual can then change and update that. But we can still see the lineage, in other words, what the AI has put in and what the individual has changed. And the value for that is that we, you know, typically men tend to be more effusive in a CV than than the women. And indeed, you know, people from a majority ethnic group tend to be more effusive than a minority ethnic group. So you tend to see a sort of leveling up in terms of the skills which is great for diversity. But then for the individual, you get real personalization, so you can understand the skill gaps against your current job, and against anything you aspire to do. And that can pull you across into the learning experience platform. And again, there we've made a very conscious choice to prioritize the skill gaps against the current job, because we all know that if you take that learning course, but then you don't actually apply that learning that learning is relatively quickly lost. So for the for the individual, it's about uploading, it's about making it a core currency, it's about how they find jobs and anything on you know, our internal jobs board. But it's also about their understanding of how they sit against the labor market, and then giving them access via Alex P to not only the 13,000 or so courses that we have created and curated, but the 10s of 1000s of courses that we have available through other providers, and then personalizing that for them. Because you can imagine if you've had got that number of courses, it's overwhelming. So it's about helping the individual a bit like Netflix to get to the the solution that they're looking for, and empowering them. But being really clear from an organizational perspective. We want to prioritize learning for the job out which skills were attached to which jobs, etc, etc. So that that's what we're seeking to do. The challenge is proving the technology needs to work. So we're piloting that. And once we've proved that, then we can talk about the use cases. And that's really important, because we've got, we're an industry which has, you know, safety critical roles as well as highly commercial roles. And so, discipline by discipline or business by business, we're building out that confidence that we can adopt that approach and switch off other solutions, which the businesses have put in place over a number of years to do those type of things often on very particularly the use cases like mentoring or some aspects of gig assessment or other things. So I aim is to create more of a unified platform.
Alexandra Foster 14:26
So interesting, sounds like you are maybe you know, a couple of laps of head ahead of where we are internally in terms of getting to the phase of piloting and testing out any girly learnings there that you're finding.
David Doe 14:42
I think some of the challenges are for example, if you've got an external course provider, they often don't come with course content descriptions. So then people immediately say, Oh, yes, let's use an AI tool to do a transcript to create and the problem with that is again, you then get a certain number of characters, it doesn't always turn up, you know, the AI generated course content doesn't remove the responsibility of someone looking over it and saying, Well, it wasn't that great to start off with, or perhaps it starts with this is blended learning. So then you immediately get a skill associated with blended learning. So you can't immediately trust just AI tools to do this. There's a there's a significant amount of investment. And there's a significant amount of skepticism. So this is as much I would say, a, as a technology journey as it is a change journey. I mean, I talked about the difference between the skill on the competence, but going through that journey, and then getting clarity there. And then making sure that you're adding when you're adding certain taxonomy into into the overall taxonomy. So for example, if I was to say subsea, the AI thinks diver, but actually we're talking about when engineering so. But actually 70% of the initial matches were accurate. So it really is about investing your time and thinking about this in a change, you know, how you do change management. And each time you roll out, we're thinking about which personas are in scope. So if someone's, you know, early on in career, they're often looking for a mentor. So how can we have a campaign that is associated with that category versus someone who's on an offshore rig? Who's really thinking about? Well, you know, what skills do I have for the job? Or how do I fit in the broader industry? And am I employable, etc. So each persona has different interests for the the sort of different aspects of the functionality. And so it's not a cookie cutter approach to change management either. That's those are some of the learnings overture,
Chris Rainey 16:43
Sarah, love us for you to jump in here. But I can see you in not in a way. I'd love to hear your thoughts.
Sarah Nutolo 16:50
Yeah, look, I because of course, what David said, actually, was talking a number of topics in the me so one around that, you know, I do, I completely agree with him on the taxonomy, coming from skills to development to jobs, which is exactly what we're facing as well. So skills that is one of the let's say, the founding, and sorry, the ontology is one of the founding areas of the system, but also one of its limitations if you like. Because if you want to have a great system, you have to define skills in advance, because otherwise, the AI won't recognize them. And therefore you hire for something, you develop for something and you grow for something else. So there needs to be a consistent taxonomy there. This is number one. But the second, which I was also curious, maybe to hear from my colleagues on their approach is, how do you bring it to life and whether in your experience, also piloting versus ailing early adopters, is the way because in my experience, piloting might not be the best way, because of the, of how the AI works. So normally, you, you know, as exactly as David was saying before, AI and skills together are reshaping the way we have change management. Because the normal approach would be you have something to implement, and you start small, and you start a pilot. But actually, with AI, this might not be the best way, because you then introduce bias into the system. Because you want the system to replicate something that might be good in a small space, but that might not apply to my bigger and wider space. So actually, in my experience, the best way might be early adopters. Because then you involve people with you in the journey, you just scale it, you ask and you allowed the engine, the AI the tool to learn as you go, and then you involve people as you go, but happy to know what my colleagues experiences about it.
Loïc Michel 19:06
Maybe I can share with you like customers examples on how they do when we work with a skills technology or talent experience platform rollout. And indeed, we see different approaches, the piloting approach is often a good one, but you need to be able to define what is good scope of the pilot, including the AI part of things. So you do not select one particular job and build your one system around this function of vertical because then you will have some discrepancies between these and the rest when you want to go at scale. So usually you select something a little bit more trunk trunk or reasonable. Then you need to have the sufficient level of data for sure. And then the sufficient engagement. So that's where probably You can have even ambassadors during your pilot. That being said, some customers do this way. So we work with companies like total energy, and that's the way they're selected to, to roll out. So stages after stages, populations after populations. On the other hand, you can have a big bang approach. And like the railway company in France, SNCF 150k users, they launched in one day, so no piloting at all, but ambassadors, and lots of change management and the connection of new ways of working with AI and new use cases on the platform and old fashioned ones. So the like, the more related to performance with an annual system to review performance. So they needed to combine that so that they had a buy in directly from everyone, because that was answering the standard approaches. And I was fueling the new approaches at the same time. And if I, if I was asked to select one good way to do it, I would not know, as a tech provider, I would go for the full rollout in one day, we are about to launch on 50,000 users in an engineering leader in Europe on Monday, in one click, that's awesome. Like the world company will be alive in one day. So that's, that's great. It's not always the case. So when you go for a pilot, you need to select really precisely the populations and the scope for sure. We talked about the our, the AI and data and digital teams are naturally a good fit for that we have lots of customers doing that, like we work with Jeeva Nova in the US. And this is for these jobs in particular that they want to start with this case based approach. So careful selection of the pilot scope, or one the big, big bit like big bank are two cool options, I would say.
Sarah Nutolo 22:00
Yeah, indeed, I probably didn't do it. Because also what I've learned that through the process, so is that really implementing technologies, what specifically when they are skilled, so which is something super new, and AR related, is really not for the faint hearted. Because you have to train people train the organization and train the AI simultaneously. And you know, going live with no chance of little adjustment beforehand. So this, I think, is also something that, you know, it's it's a great, it's a great challenge, it might also be a bit of a limitation, depending on the risk inclination of the different companies, but it's definitely an element to consider.
Alexandra Foster 22:48
Yeah, well, let me think you also have to Sara, think about the different personas, right? And so you say, where do you get the best value from an AI perspective? Oftentimes, we think about testing things out and piloting them, sometimes we say, let's give it to our top performers. Let's give it to, you know, the strongest members of our team folks who've been enrolled, that are excelling. And what what our experience has been is that you need to take a more representative, and, you know, inclusive approach, because sometimes the greatest value is derived from those individuals who are yes, they're performing on target, but maybe they could use to help. I'm working at a higher level, completing some of those tasks. And so I think, our teams, our AI team has done a good job of when we are piloting and testing things out to be very intentional about the pilot groups. And to really understand what we're measuring, what does success look like? And it is not just the deployment of it, obviously, it's the impact and the value, the value to that individual value to the team and the value to the organization. But I do think it's a really great, a great challenge to think about how do we do that in a way that is it isn't throwing things off in the organization? Because it is a lot of change at one time.
Chris Rainey 24:07
Yeah. So talking to change, Tony, what what are some of the things that you've got in terms of the ethical considerations? You don't want to? I don't think we can have a conversation around AI and not talk about some of the ethical considerations on implementing this in talent management. How have you addressed?
Tony Wiseman 24:25
Yeah, yeah. Thinking about that, Chris, as the guys were talking and yeah, like, you introduced me earlier as the digital HR director in Zambia. And I also have for my sins been asked to lead a group to develop our ai ai policy for the whole organization as well as the employee guidelines. And, like one of the things that's come up again and again, with regard to policy and guidelines is the ethical side of it. I just wasn't being an HR American an HR For so long, and when we're dealing with, with personal data, you know, and the use of personal data, ethics is always a big piece. But no more than, than ever, it's crucial. And there's an acronym that I use. It's a BT PA. And the BT P stands for bias, transparency, privacy and accountability. And when we talk about it, the be the boy's voice and discrimination, like the real danger, I suppose you have with AI systems is that they can perpetuate existing biases, if they are trained on historical bias data. So really important that kind of when we're looking at the data that you're using diverse data sets, but also I think it's really important. And this is, I suppose, where we need to engage more with the industry, technology industry, as well as how do we audit AI algorithms for bias? You know, that that's really important. Second piece is in transparency, I think one of the things that we've learned is, how can we gain trust with our employees, and one of the things is really is trying to be clear or transparent, and how AI makes its decisions, particularly in the talent management space, you know, because this, this is going to impact people's careers or people's lives. So trying to be transparent and how it comes to these decisions is really important. The PII the privacy. There, this is, for me, it's quite it's, it's, it's so important, but it's difficult as well, because there's so many different jurisdictions, and how we ensure compliance with data protection laws, and all those different jurisdictions is something that I suppose we struggled with, particularly with the onset of generic generative AI. So we use an AI I suppose for the last number of years generative AI, is, you know, brings a whole new set of problems in that space. And then accountability, in terms of my BT PA, but accountability. I think the guys have mentioned this already. But it's so important that there's a human in the loop and all of this so that if a decision, AI comes up with a decision that someone can, you know, review that decision, and that decision can be reversed or changed. If you're not happy with that accountabilities, and someone needs to be accountable for the decision and can't just blame AI. I mean, that's really, really important. So yeah, that's kind of my take on the unethical side, that BT PA and looking at it from that perspective,
Chris Rainey 27:39
I feel like everyone's having to go through a similar process, right now in building that out in your organization. So
Loïc Michel 27:47
from a from a provider standpoint, you know, because we are providing the services and the tech and the AI. And we are responsible for the impact on dry, we need to be we need to be really relevant and really like, at the top level on all these BTP approaches. So that takes different, different angles. But you know, like, as a European leader in the space, we need to comply with the new regulations coming on AI, the ACI act in Europe, which is going to be published in a couple of weeks, then you'd have applicable texts for the different areas, including HR, HR related areas. So you need to be like benchmarking, getting advice from the experts and working on that. Good news is that big organization like the ISO are launching some new ways to approve some of the AI management systems. So there is a new ILO certification, we are running at risk five times, which is the ITIL for two or one, and it's brand new. So I guess it's important to lie to this kind of organization that validate the systems because decades ago, it was all about quality management system. Then we had security management system and cyber security, you had sock two, then we had GDPR, for the privacy for the PII of the BDPA. And now there is AI so it's good to see that you have some like top down approaches and frameworks you can rely you can rely on. So that's something I guess it's important to ask to a provider when you are you are meeting some of them a how do you comply with these regulations? And how do you translate this regulation into applicable measures into your product and into what we need to do with the product?
Speaker 1 29:50
As a provider, do you see yourself getting asked those questions a lot? Because you know I have asked questions too. different providers at different times. And sometimes the answer you get back probably isn't what I'd expect. And just Yeah, I'm just wondering in terms of, like, you know, if someone asks you, you know, can can we audit your AI algorithms to check for bias? How would you even go about that? What will be the approach? Is it because you would you just show them that you're meeting these standards are meeting the legislative standards?
Loïc Michel 30:23
Yeah, I think they there is a, like, some conviction on like, the DNA of the way you were doing it, the product itself, and we talked about skills based talent management and talent experience, I think, if your system is really built around skills, then you might have some discrepancies in the way skills are detected or expressed by the people. But that's better than anything else. If you look in the mirror, like in the mirror and in the, in your back with historical data, and you try to replicate, then you will replicate the bias with the skills data, you have this common language that that is, the less bias possible, maybe it's not 100%, unbiased, so you need to validate that. So first, like the approach the DNA of the product, how it's built. And then you have Yes, this big concept of transparency, responsiveness, responsibility, and Explainable AI. So as a provider, we just put our AI experts in front of the customers in front of the HR in front of the users in front of like the employees, the managers, and they do some keynotes on AI and the impact on HR, and why is doing this and not that, so that it's even training the users so that everyone is upskilling in that space, then we have some documentations, as well. So we have the AI fractured by three, six lifetimes. So it provides a lot of information as well. Jenny i is accelerating requests from customers. And so it's really evolving. And then they say, oh, Jenny is is a new thing. How's it working? Now we need to comply to this and this and this regulations internally? Can you validate that it is not that I'm not doing this? And then we just saw even with existing customers, like large banks, or global banks, for example, they've just said, Okay, we need to stop, not program the project, but the implementation of these new capabilities in the product, take a breath, take a few weeks, validate some elements, and then we'll give a go to the full the full implementation and that what happened to just ask a little bit more work on our end, that because we have all these different answers, we are able to comply, and then to reassure and then to go ahead. Yeah.
Chris Rainey 32:49
Also, the pace of the change is moving so quickly that you haven't to update those documents every day. At the moment, right, every every every few months go by there's a new release, that there's more updates more considerations. Right. I had a question from David that popped up earlier, David, that someone asked, you obviously walk through sort of the rationale behind the change that you're going through and the technology roadmap, and someone asked, how has that changed? Or what do you do with your team, as that freed you up? And you and the team up to do other types of work? Has it changed the way you know, just just out of interest?
David Doe 33:29
So as I mentioned, we're at the stage of piloting. So we we will actually only really go live in August. Oh, wow. Okay, there's an awful lot of work that needs to take. I mean, when you look at an organization of our size, and the number of different job profiles we need to ingest, there's quite a lot of cleanup work. And then it's the data in and the data out. We've had to be involved in quite deep conversations on ethics, etc. In the longer term, I can absolutely see the return the ability to understand supply demands, the ability to step away from individual conversations around job opportunities, and actually make things more transparent. But the reality is, this is a complex space. And so it doesn't come without upfront investment. And I've tried to be clear on that earlier. And really, you know, ultimately to follow on from the points that have just been made. Unfortunately, AI is in a position now where there was a an Austro, Austrian, Anglo, I think philosopher called Michael Pollyanna who said that basically humans know more than they can tell, well, now computers know more than they can tell. So ultimately, you know, you can't have a there is no transparent system. If anybody turns up and says, I can tell you how the AI is made a decision. Deep learning and certainly generative AI can't do that. So you have to be really clear on what fields we're giving the data access to. And so All of that type of stuff you need to do. And then there's a huge amount of change management effort. So there is a substantial payoff, but it's not up front. And AI does not work perfectly out of the box. So having to explain to colleagues, they say, oh, no, I don't need to do all of that AI will do it for me. Well, you know, have you seen the quality of our job copy, you know, job descriptions. Do you know the average person may not be able to understand some of the job titles, I don't understand some of the job titles. I've been in the company 21 years. So. And I've worked across a number of our businesses, so you can imagine the amount of effort that needs to go in. But once you've done it, as I mentioned earlier, things get much easier, much more rapidly. So and it kind of speaks to Sarah's point about taking a sufficient size and sufficient cross section. But it's building that confidence each time because if you switch something on, and then people say it's rubbish, you lose people at the get go. So, yeah.
Speaker 1 36:07
That's David's message. And it's so important, isn't this, you know, like, because when people hear you're doing something on AI, or gene AI, it seems it's gonna solve all the problems of the world. You know, if you're not,
David Doe 36:20
it's a lot of hype here. And there's a lot of hype around, we're gonna get rid of traditional jobs. No, we won't. Our entire legal infrastructure is built around traditional jobs, skills are very helpful, they're helpful for agility, they're helpful for personal development, they're helpful to understand shifts in the market, they're helpful for personalization, they're helpful in order to rationalize your learning portfolio, to speed up, your time to fill, etc, etc. But the idea that suddenly we're gonna abandon jobs I'll come on, it's not reality, it's been presented. It's one of those things, which is an interesting thought exercise. And a lot of people are talking about it, but I don't believe it. And I know, some organizations may have that because they're consultancy type organizations. But if you've got an organization that has assets, you know, that sunk capital, you're going to have traditional jobs for quite some time, skills are just going to be part of that as they have been for a very long time. But we've we've kind of done that by, you know, talking in a way where I say your skill and you you think you understand it, but we're talking a different language, the AI now is a means of basically creating out what I would call a Rosetta Stone, so that you really understand that you're saying the same thing. And you've got some understanding across the organization. And you can help workforce planning and a whole host of other things. But it's not some form of Nirvana that will arrive overnight, or at least that's my
Chris Rainey 37:49
belief, you just shattered everyone's dreams, David, you just tuned in, they're like, they all came in thinking this is gonna be No, yeah, no, I love it. And this is why it's important to have these conversations because I see some, definitely companies entering into the space, not realizing all of the things that you're sharing that everyone's just shared right now, they don't really think that they have this new shiny technology, and it's all gonna be great. And they then realize, actually, it's more of a cultural transformational challenge. And change, as as much as it is a technology. So it's really important that we that we come together and share,
David Doe 38:22
Chris, I mean, I think if you don't do this, and others do it, and others are able to adapt, if you think that, you know, in five years time generative AI is likely to change jobs, quite substantively. And you don't understand that, then I think you're at a disadvantage. So the challenge is, is the technology in a sense, you when you jump on this, and you jump on it in a responsible way and ethical way, in a way that fits your business that meets your use cases. But if you turn a blind eye to it, I think in five or six years time, you will find that you are no longer as agile as your competition. That's the challenge.
Sarah Nutolo 38:57
And you're very right to David. And I actually believe that in this journey towards the future, this is what the the biggest advantage of looking into the world or from a skill lens would bring, because they will bring something that the traditional jobs won't be able to bring. So I agree with you. So we will never get rid of jobs so not really in the near future. But definitely if we look in a future where in 2030 for the one of the latest Deloitte research tells us, the number of people 65 years old will outnumber the number of people below 18. This means that that talent shortage is a reality that happens tomorrow. So the point on skills combined with AI will be Are they helping us to accelerate this transition and how? And for instance, by looking into skills and skill adjacencies Are we looking into a way of bringing all of it to gather, and for instance, starting to tearing down the traditional function definitions do we will need function in future? So for me, the question is no more do we need jobs in the future, but do we really need functions in the future.
David Doe 40:15
And I think, you know, you're also going to see, to your point with generative AI, the the interesting thing here is it, it lifts people of lower ability up to do tasks that they haven't been able to do before. So your ability to then widen your talent pool, to at a time when business models are changing when we have, you know, uh, you know, the world of work has never been changing faster than it is today. And we have a challenge around meeting climate change and an aging population. So the aim of getting productivity, overall up getting more people into the workforce is is, you know, something that we have to rise to. And ultimately, this is an enabler as a technology in the long term. But it's finding your way into it, and how you want to use it and not seeing this as some form of, you know, unicorn solution, it won't be. And to your point, I think, you know, how we think about this, and how functions, at the end of the day, the aim should be to empower line managers to put this data in front of them, so that they don't necessarily need very high overhead. I think, you know, HR will still be there, but it will be different to what it is today as it was 10 years ago. I mean, when I was 20 years ago, when I first came in to shell I was sitting down with people and thinking about whether I gave them an extra 50 pounds on their on their pay increase. And we don't do that anymore. And we haven't done that for 15 years, and in 10 years or 15 years, we won't be doing the things we are today. So technology will be a key enabler that always has been.
Alexandra Foster 41:50
Yeah, I think the key that I that we need to spend our time on is staying ready, right? Because it's ever evolving. And and I think your note, David around turning a blind eye to it only disadvantages you in the future. And so I think it is about understanding at a core. What's your perspective from from an ethics standpoint, which are cultural readiness? And you mentioned something earlier, just like what's your readiness? So you say you're ready for AI? But are your systems today? Are they ready? Are they consistent? Do you understand when you have a skill and in a job description? Is everybody on the same page there? So I think there's work that organizations can do now, even before they begin to scale up AI operations. So it's energizing
David Doe 42:37
to pay on knowledge management, and a whole host of other things and, and generate or create other problems, performance management, how do you how do you? How do you track that? You know, it's gonna create EIR challenges that we haven't had in a long time, you know, change your recruitment strategy. So, so much will change. And I think it's about, you know, it's about how you understand that you have the data to be able to adapt to those challenges. Yeah,
Loïc Michel 43:02
David, on this one, on particular, I think where technology can really help is in building the first layer, like the foundation of all these transformations. And you mentioned the Rosetta Rosetta Stone. So the common language that you can create with the technology right now, I think it's really important to put that image to the maximum. So Rosetta Stone means being able to work in as many languages as you would like. So that's something I think in your global organization, it's really important to have a really clear focus on our do we make sure we have the same approach globally. And we are not separating the approaches just because of the language because we are not able in different countries and cultures, because we are using different languages to do the same thing in terms of HR. So I think that's a really precise area where an AI driven skills technology should be able to help you. If it's not, I think it's a problem. And I think you need to look for probably from other providers and to challenge your existing because that's key in your in your global scope.
David Doe 44:11
I agree. And so many people say, Oh, I've got the gold standard of a taxonomy it doesn't. So ultimately, it's interoperability in the Rosetta Stone analogy of how you are able to understand what in one language is this and another language is called something slightly different. But that's human culture. Right. Therefore, you I mean, that's, that's, that's diversity at its core, right. Yeah. That's the classic around why certain cultures called one thing one thing and you know, and even in the English language that that happens all the time. You just need to cross the Atlantic divide.
Chris Rainey 44:44
Yes. Yeah. Wow. That was a lot to take in. And I thought we could do a whole series. Each of these things. I mean, one of things I was thinking about what Sam was talking is, AI in many ways is like a great equalizer. Right? And I I think someone mentioned earlier I think was was mentioned earlier.
Alexandra Foster 45:06
I think we assume it's valid me, I think,
Chris Rainey 45:08
yeah, I just wanted to remember that because because we're now able to do some of these things at scale that we could just never do before, right, empowering our employees to take charge of their careers, right? Is it my day, I'm not that old. But I mean, I started when I was 17, you didn't have that opportunity to to really take it into your own hands in a way you can do now in terms of upskilling and rescaling and being able to tap into a talent marketplace and apply for opportunities, etc. So employees are going to now feel really empowered to take charge of their careers, as well. And, and to Laura's point. Now doing that globally, because AI can translate into, I think it's 97% of the world's languages at this point. And in two months time, you'll have voice with the release of open AI voice capability. Now you can talk to the AI in 97% of the world's languages, and have a conversation, as well. So this is just the beginning. So if you're not on board, now, you need to get on board, you're already behind. Even a small company like ours with 15 employees. It's been we've been working in this space for three years building our and our copilot of HR leaders, because we haven't we realized as content creators, that we're going to be disrupted, as well. So epic, no, no, there's no function industry, or anyone who's not going to be disrupted by this, in my opinion, as well. So this exciting times, I've been told by the team, but we have to wrap up. But before I let everyone go, like pining piece of advice, so what would be your parting piece of advice for everyone? And then and then we'll say goodbye, Sara gonna kick us off? Yeah,
Sarah Nutolo 46:50
absolutely. So I think the piece of advice would be to look into the user experience. First, whatever you do, be let user experience be your light, because it will drive adoption, it will drive consistency, it will drive skills, and it will drive a better learning for the tool as well. So if there is one thing for you to concentrate on, let it be the user experience.
Chris Rainey 47:19
Tony, yeah, I
Tony Wiseman 47:20
think get started, you know, get started with me. My drivers got going. Don't waste any longer. Don't wait for the perfect time. Amer perfect technology just get started.
Chris Rainey 47:30
All right, Alex, I'd
Alexandra Foster 47:33
say keep humans at the center of what you're doing. Right. I think delivering on that experience that Sarah shared, but also thinking about the long lasting impacts of our decision will lead for generations to come will be critical.
Chris Rainey 47:46
Yeah, super important. We can't forget that part, right? Like, yes,
Loïc Michel 47:51
I will probably be stealing the Rosetta Stone from David, look for your Rosetta Stone, ask your tech providers to be able to build that Rosetta Stone. And it's not that complicated to build. So if someone is telling you that you need six to 12 months to get ready. That's not true. Because if the product and the technology are well designed, and if the company is understanding your business and your way you do things, then it could be done in a matter of weeks. And that will be the that will be the short track I will be looking for so probably probably agreeing with Tony as well. On The Go Go for it. Yeah, yeah,
Chris Rainey 48:34
you just put all of these tech companies under so much pressure right now. They're all like said you can do in two weeks.
Loïc Michel 48:45
This is what we can do to some extent, of course, we can adapt to some contexts that we did that in eight weeks for really global organization on 1000s and 1000s, and 1000s of users globally. So
Chris Rainey 48:58
it can be done. Everyone's sweating. Now. Last one, not least David.
David Doe 49:03
So a lot of it's been said, but I would see it as the key to helping you to be agile. Don't see it as a pre deterministic thing around skills, it's going to tell you completely the future. If you went back 24 months, everyone would have been a prompt engineer. And now to your point, you can talk to you in two months, you'll be able to talk to generative AI. So this is about navigating a world that is changing. And you a certain moment work out when you want to get on the train, how you want to get on the train. And and there is no cookie cutter approach. You need to pick the thing that works for you in your context.
Chris Rainey 49:39
Yeah, and the train is going to get faster. Keep moving. So jump on now. Well, that's it everyone. Thank you so much for tuning in. massive thank you to all of our panelists. Thank you all for sharing and joining us and for everyone that's tuned in from all over the world. Obviously special thank you to our friends at 365 talents for helping us bring this panel together wherever you are watching right now there's a big giant button. But beneath the video that you're watching, if you click that you can download their their latest report, which actually teaches you how to leverage AI and HR to help shape the future of work. So there's a very practical steps you can take that download that it's free. Also, as many of you you know, we launched the HR leaders Atlas co pilot a few weeks ago. So you can also download that in the chat that's been trained on over 10,000 hours of content that we've created. Thank you again, everyone, wherever you are in the world, and I wish you all the best said everyone. Thanks bye bye
More from the HR Leaders Podcast
Victoria Klug, HR Director Eastern Europe at Beiersdorf.