Joe Fuller: If there’s a new bargain between corporate employers and employees, skills building is at the heart of it. Pay, benefits, and company values still matter to workers. But career development and opportunities to advance are essential in a fast-changing digital economy. Conversely, employers seek workers who are motivated to learn. And they commonly evaluate them on their skills development as well their business performance. In light of this mandate, supports like flexible work and recognition of caregiving obligations take on added significance. How is this playing out and what is the benchmark for HR strategy in the future?

Welcome to the Managing the Future of Work podcast from Harvard Business School. I’m your host, Harvard Business School professor and nonresident senior fellow at the American Enterprise Institute, Joe Fuller. I’m joined today by Nickle LaMoreaux, Chief Human Resources Officer at IBM. We’ll talk about how the iconic tech firm’s approach to HR has changed over the past several decades and how AI is transforming the practice. We’ll also talk about the company’s skills strategy, from skills-based hiring to training and reskilling and how AI helps tailor approaches throughout its quarter-million-strong workforce. We’ll also consider how AI is likely to alter work and jobs going forward. And we’ll discuss how adopting AI is as much about company culture, values, employee experience, and talent management as it is about technology, performance, and business results.

So Nickle, welcome to the Managing the Future of Work podcast.

Nickle LaMoreaux: Thanks for having me, Joe.

Fuller: Nickle, you're the CHRO of one of the most iconic companies, certainly in the United States, if not the world. How do you find yourself in that position? Tell us a little bit about your journey.

LaMoreaux: So many people would call me a lifer. A lifer in many respects. I have spent almost 25 years at IBM. I started as an intern in human resources. I've been in talent acquisition and compensation, learning and development. I've had an opportunity to do two international assignments in China during my career, so to get different geographic perspectives. And I've worked across a variety of our businesses, hardware, software, services, sales. And so I think it's been really being able to go deep in some areas, as well as going broad that has landed me in this seat.

Fuller: So as a lifer, you will have gone through multiple periods at IBM, periods of retrenchment, period of growth, major pivots in strategy. How has that shown up in the evolution of human resources policy and practice at IBM?

LaMoreaux: I think most companies have gone through three, some would argue we're now entering kind of a fourth phase of human resources, we call it. HR 1.0 was the area of compliance. Personnel departments were mainly focused on compliance, whether it was around payroll or employment. And then we moved into what I'll call was HR 2.0, where companies started to focus more on process and standardization. Some of this was focused on the employee experience, so that as organizations grew, you'd have consistent experiences, but it was mainly focused on efficiencies and productivity. And we stayed there for a long time. This is where centers of excellence started and policy standardization. But then many companies and we moved to what I'll call is HR 3.0. And we probably started this journey in 2015 to 2017, and that's where we started to focus more on the employee experience. So with standardization, with processes that were making us compliant, how could you also make sure that they were seamless to the employee, that they were easy to navigate? And all of these build on each other. But now with the advent of AI and technology, I think we're entering it into HR 4.0, which is allowing us to do customization at scale, personalized experiences at scale. The career advice you may get may be different than the career advice I get based on our personas and what we've done. And so I think we're now in this interesting time, and that's the evolution I've kind of lived through at IBM.

Fuller: Those different stages suggest an evolution in the way you're defining the relationship between the employer and the employee, how that is thought of by both parties. Does that distinction make sense to you? And how would you describe the type of relationship IBM's trying to form with its employees in the 2020s?

LaMoreaux: Yes, I think that is exactly what you would say is that evolution of HR also changes with that relationship between the employer-employee and that contract. Maybe one point I'll say for the listeners on this evolution of HR. You can be in multiple states at the same time. So I don't mean to say that just an HR department is very singularly in one state. And why I think that's important is the employee value proposition or what employees are looking for from their employer has gone through a similar transformation. As an example, I think there was a period of time where the employee-employer contract was really based on terms and conditions. This was a period of time where you would join one company. You would probably stay for your entire career. And that's when HR departments probably focused most on compliance and most on standardization. Then when we got to a point where it was no longer the norm to stay at one place for your entire career and there was more employee mobility changing companies, I think what we started to see is yes, employees still cared about terms and conditions, but a second piece has been added to the value proposition that employers must provide, and that is skill building. Am I building skills in my current job? Am I being prepared for what the next job might bring? And am I building skills that can be used over a career? No longer do employees think that the skills they have today will last them a 30-year career. They know that the rate and pace of industry and technology changes. They're going to have to be prepared to be continuously building skills. So skill building, learning has become a big part of what employers talk about in recruiting processes, but also in retaining employees. But I think over the last maybe three to five years, there's a third piece, again, these are all additive, that have been brought to the employer-employee contract, and that is the values, the culture of a company. With so many choices that exist in the labor market today, employees want good terms and conditions and they want opportunities for learning and development, but they also want to work with a company and an industry that aligns to whatever their values are. And that is very customizable if you go to the evolution of HR now needing to be customizable. And so I think as employers and as HR departments, talent acquisition professionals, articulate to potential prospective candidates why you might want to join the company. You not only have to talk about the processes, the employee experience they'll encounter, but you have to talk about all three parts of this value proposition.

Fuller: What do you think catalyzed this change? Is it changing attitudes in the workforce? Is it more forward-thinking executive talent boards of directors that realize that a company could differentiate itself by embracing this refined definition of the deal between employees and employers? What's at the heart of it, do you think, Nickle?

LaMoreaux: I think there are a couple driving forces. One is the mobility of the labor force and the fact that you don't have to stay at one place for your entire career to get maybe benefits that you would've gotten in the past. So I do think mobility of the workforce, the hybrid environment, and even remote work has enabled that. So you're no longer just tied to a specific city or country to think about your employment option. So it has gotten more competitive. But I also think there is another piece that has driven this. And if your listeners have seen the recent reports on the trust index [Edelman Trust Barometer] that was published by Edelman, corporations are now viewed as the most trusted entities above academic institutions, above religious institutions, above political institutions. That shift has allowed potential employees or existing employees to have different conversations with their employers about the role we are expected to play in society and having different points of view on various topics.

Fuller: Well, I'm sure activists from the '60s and '70s are absolutely astounded by the results of that Edelman survey. A graduate of our school, Mr. Edelman. Let's jump to a couple of the other issues that are being widely discussed related to employment and how companies go about attracting and keeping talent. A major movement in the last couple of years in the US has been a shift to what's called skills-based hiring, where companies are trying to move beyond credentials like college degrees as proxies for evaluating candidates and are really trying to understand what skills those people bring as opposed to what credentials they've got that suggest they have skills to the workplace. IBM was an early mover in this space. How are you approaching that? How have you observed this broader shift to skills-based hiring? And do you think it's actually making material changes in who gets hired? Because while having a corporate policy that says you must have a college degree to work, let's just make it easy, at Harvard Business School where I work, if you remove that, it doesn't mean the hiring manager or the person that's actually going to pick the candidate is going to opt for the non-college graduate if given a choice between the two. So tell us about your journey in skills-based hiring, where you think we are now, and what the prospects are.

LaMoreaux: So we are huge proponents of skills-based hiring, not just because it gives you a broader pool of candidates. And in this labor environment, particularly in the tech space where there are massive skill shortages, this isn't just a numbers game for us. We actually believe that this leads us to better employees, better candidates, better hires. So for us, this idea of skills-first is something we started in 2016. What we started to find is particularly in technology, and then this could even be broadened to others we found, but in 2016, what we found in technology is that we saw many candidates who were getting technology skills in non-traditional ways. And so if you then say that you must have a college degree, you are actually taking a large portion of the workforce that could be qualified and automatically excluding them. The latest statistics show that 62 percent of American adults do not have a college degree. And so in this labor environment, can your organization afford to just unilaterally exclude 62 percent of the potential applicant pool? And when you go to some diverse groups, it's actually a much larger number that you're excluding. So for us, this is a skill statement, this is a quality of candidate statement, and this is also a diversity, equity, and inclusion statement. Python is a technical skill that we recruit a lot for. Why do I care if you learned Python in a four-year college degree, or if you took it at a class at a community college, or you learned it in the military, or you taught yourself online at night? I don't, as long as you have the skill and you can code in Python. And so that's a really just practical example that you want to make sure that you've got fairness in your process, but also that you're getting qualified applicants. Right now, 50 percent of our jobs in the United States do not require a college degree. Clearly there are some job roles that that degree requirement may stay, or a legal team will need to have a law degree, of course, but there may be some jobs underneath it. Even if I think about HR, talent acquisition is a good space where some of our most qualified candidates actually don't have a college degree, but instead come from the retail sector where they've had lots of experience with interviews, with application processes that translate very well into the talent acquisition team. So I really encourage all organizations, I know that college degrees are important, and this isn't a college degree versus not having a college degree. For me, how I learned my skills was a college degree. And so every candidate will be different. This is about opening the aperture of your candidate pool to make sure it encompasses all qualified candidates.

Fuller: Another one that's come up really I think provoked by COVID or moved to the front of the queue by COVID, but was something that we were studying in our Managing the Future of Work project prior to COVID is this whole notion of how caregiving obligations make their way into the workforce with things like higher voluntary turnover, especially in high-paid jobs, presenteeism where someone comes to work, but their head isn't in the game. They're worried about the call from the doctor, the call from the school principal, or they've got parents where one has got a dementia or an Alzheimer's diagnosis. All very familiar stages of life and stories we've all lived through personally or lived through with friends. How are you thinking about accommodating that at IBM? The old deal that we were talking about earlier was one in which you got paid, you had some benefits, you had friends at work, but your employer not only didn't have anything to do or say about your child with a chronic condition or the distraction of taking care of your parents. How are you tackling it? Do you see it as a major issue in terms of productivity, turnover? And also, that engagement, is it part of the new deal at IBM?

LaMoreaux: Yeah, it is part of the new deal at IBM. There were many difficult things during the pandemic. But one of the good things to come out of the pandemic when people's personal lives and their professional lives collided in ways that you never expected, right? Home was now your office and it was your home. And in some cases, it was the schoolhouse for your children, right? So it was all of those things. Managers and organizations had to very quickly adapt to what we would call empathetic leadership, listening, understanding, knowing when to probe deeper, when an employee was not ready to share more. How could you make the work environment more suitable for them? And so I think that those fundamentals we are training around empathetic leadership. The second thing we're doing is we believe in flexibility. We do have a hybrid approach to work for most job roles. That means three days a week in the office. There are some job roles where it's five based on the type of work they do. There are some job roles that could be fully remote. But for most of our employee base, it's three days a week in the office. But what three days a week those are get to be determined between you and your manager. The second thing is the flexibility around time. So even if you need to be in the office a day, can we be flexible around hours that accommodate some of your responsibilities outside of the workplace? And again, these aren't top-down mandates. This is between an employee and a manager. And I think that flexibility helps balance the professional and the personal obligations. The last thing that we do is we do acknowledge that even with these first two pieces, that for some employees, taking time away from the workplace is the best thing for them. And so how do you have leave programs that allow people to take that time back, gracefully come back if it's shorter. But also we offer a program as an example called Tech Re-Entry. This is targeted at people that for various reasons may have had to take 5, 10, 15 years out of the workforce. And as we talked about earlier, given the half-life of skills as shortening, they may have been in tech 10 years ago, but tech today looks very different. And so as we re-recruit these individuals back to the workforce, we give them a training program, an assimilation program in the new technical skills that are needed knowing that they have some foundation in technology. I think all employers need to think about all three of these components to the what I'll call the “cultural piece” of empathetic leadership, the very tactical operational about how you put flexibility in your workforce. But don't underestimate the power of some of these longer-term programs like Tech Re-Entry.

Fuller: Well, that's a great illustration of how a company can increase its what a fisherman would call catchment area of talent by creating some customized pathways to take what our research calls hidden workers that would ordinarily not be considered for employment and create avenues for them to reenter the workforce in the case of career break people or overcome other issues, everything from neurodiversity to creating pathways for veterans whose skills are usually overweighted on the soft skills, underweighted on the hard side. Let's go back to something else you mentioned. It's certainly absolutely redolent in our research that employees judge a company on the learning environment it creates. Now, of course, given the sectors IBM competes in, you're at the epicenter of this ever-accelerating technological perpetual motion machine we've got. What kind of role is HR doing to take a structural approach to re-skilling as opposed to relying on traditional on-the-job training or self-motivated learning by people just interested in technology, the alternative, more traditional ways it was dealt with?

LaMoreaux: Yeah, and I think this is a fundamental question that all organizations, all HR departments need to ask themselves and really need to get right for not only just short-term growth, but I would even say long-term viability. You've got to have a workforce that fundamentally and an organizational culture that fundamentally supports continuous learning. We do selection, and one of the traits that we look for is this concept of continuous learning and learning agility. So just like as an organization, you might look for some soft skills and soft traits like growth-minded, as an example. We are testing and we are interviewing candidates on these two very important capabilities. Inside the organization then, this is not just a do you have it, do you not have it? We have to make sure we're living up to our end of the bargain and providing employees an opportunity to learn to reskill in an easy way. So there's a couple things we've done. We have a learning platform called Your Learning. I like to call it the Netflix for IBM learning. It's an online platform. It is where all of our learning content is housed, not just where we can give you specific training that we want you to take for your job role, but where you can also get training that you need, that you're interested in. Just as you said Joe, people are thinking much more broader. They're thinking beyond just their current job role. They're thinking, how can I do something else that might be adjacent or might be in a totally different field? So this platform will actually serve up for you learning recommendations based on your profile, and it will create customized, using AI, roadmaps. So if you and I both went into the platform and I said I wanted to be a Python developer and you said you wanted to be a Python developer, rather than giving us just one generic roadmap of here's what you would need to do, this platform would create a customized roadmap for you. And it might say that you need only 16 hours to become a Python developer and it might tell me I need 1,600 hours. But it knows the classes we've already taken, the certifications we already have, the job experiences, and it's giving us that. And I think making simple, easy learning opportunities for employees is important. You also have to have a culture where learning is expected and rewarded. So we have a performance management system that has two dimensions. First one won't surprise any of you, employees get rated on their business performance. But the second equally weighted, equally important dimension is skills. So employees get evaluated on how are they building their skills, and that can be skills in their own domain deepening it, or as you talked about, becoming T-shaped, broadening their skills. And that then flows through to our promotion programs, our rewards and recognition programs, and our compensation programs. And I think that's really important. Because it's easy to say reskilling is important, but are you willing to put it at the center of your rewards recognition systems? Are you willing to put it at the center of your performance management systems? And so that's something we've done to build that culture.

Fuller: The way you describe it, it seems it's a very integrated program, that there's a clear line of sight across processes, which in many complex organizations you actually don't see. There's a learning agenda and a corporate learning officer and corporate learning capability, but it isn't really synced up to promotion advancement. It's not synced up to personal goal setting. It's more related to position and things like that. You mentioned AI, which here we are talking in right around Groundhog Day in 2024. Obviously IBM is a company not only that is using AI, as you just alluded to, but develops AI products, arguably is the company that first really introduced AI to the public consciousness through Watson and some of the extraordinary accomplishments that your technology people achieved with it. As you think about AI, what kind of guidelines are you using to adopt it? That's a debate in many companies. How pervasive should it be? How do you control it? And where do you see the best applications for AI in the human assets area?

LaMoreaux: Yeah, the primary discussion around AI that HR professionals need to make sure they have a voice in in their organizations is, this is a culture discussion, this is a values discussion, and it is an employee experience discussion and a talent discussion, and then it's a technology discussion last. So to answer the first part of your question around what have we done around policy, principle, guidelines, we as a company have come up with three principles that when we implement AI are really important to us. The first is we believe that the purpose of AI is to augment human intelligence, not replace it. In fact, I know this ship has already sailed, but I have said this many, many times. If we could go back and rename artificial intelligence, we would actually call it augmented intelligence. Because I think that as we all go through this journey around AI, what we're going to find is that's a really important principle. So for us what that means is AI will never be a decision maker in HR. We will use AI. We will augment and bring forward to our HR professionals, our managers and our employees information, but AI itself will never be decision maker . The second one is data and insights belong to their creator. From a business perspective, what that means is for our clients, anything that AI and our technology generates for them, it is owned by them. It is not owned by us. Internally for employees, what this means is that we're not going to let AI run wild.

Fuller: It does sound also being familiar with a lot of the debates about potential regulation and what's gone on in Brussels with EU and even local jurisdictions like New York that your three policies match up quite consistently with the areas that governments are expressing concerns about.

LaMoreaux: Absolutely. And if I even go one step further, because you mentioned some of those regulations and we are active proponents and many of the government discussions around how they should be thinking about AI. We also have five pillars. That anytime an AI solution is about to be used internally and certainly in HR, they are five pieces of a checklist that you've got to check off. The first one is explainability. Anytime we're going to use AI internally, is it built into the solution that it is explainable, that again, the user can click on something or see how the AI is being used and why it's coming to a recommendation that it's coming to. The second is fairness. So we need to make sure in our systems that there is equitable treatment between individuals or groups of individuals to make sure that there's no bias, right? The third one is robustness. If you have a narrow set of data or proof points that you're then using to put AI at scale, it may not be robust enough to scale across large portions of the population or certainly different countries. So robustness is one of the things that make sure that AI is fair and explainable. Transparency is the other point. Again, what data is the model using? How is it using it? And the last one is we do believe that AI systems should prioritize privacy. And so those are five things that as we implement, we are really running. And before anything can go live, it must meet those five criteria.

Fuller: So where do you see the frontier for AI broadly, but also in terms of continuing the evolution of HR with your five phases? Are we going to be entering the sixth AI phase? And specifically, where are the opportunities to unlock value through a more broad embrace of AI in HR? What are the problems you think that AI can help companies and workers and employees at those companies address as we get more comfortable with it, as training data becomes more available, as we understand how to manage the crosswalks between open generative AI systems and their smaller twin cousins inside companies?

LaMoreaux: A lot of conversation, debate, maybe even some salacious headlines are out there about AI and the impact on workers, on laborers, on employees. And I think what is important for all of your listeners to understand is, again, on this topic of I think AI will augment human intelligence not replace it, there is not going to be massive job loss because of AI. It's a very, very small percentage of roles that will be completely eliminated. I think the flip side is also true. I know there's a lot of conversation of are there going to be all these new jobs that are created that never existed before? I also think that in the short to medium term, that's probably not true either. It's going to be a very small percentage. What I think we all need to focus on is the 95 percent-plus of all of our jobs that are still going to exist, but how we get our work done, what we do in those jobs is going to dramatically change. And so I think AI really has three big use cases. One is AI can generate recommendations. We talked about earlier in the episode about learning plans. It can generate very customizable fit for purpose recommendations that then humans, because they are the decision makers, can decide whether to follow or not. So I think there is a big unlock there to make things fit for purpose, to save time, to create better experiences in the recommendations category. A second way that I think AI can unlock value in the workforce is what we might call digital assistance, questions and answers, surfacing up information very quickly. Some users have experienced this in chatbots, rather than trying to find information, searching through web pages. You need an HR policy? What is the vacation policy? Don't go set up a meeting with an HR professional or don't go try to find it on the web. Ask a chatbot. And so I think that that is going to surface up information much more quickly and create better experiences in our organization and have more consistency of answers. The last piece of this, and I'm pretty excited about the last piece of this, is what we might call intelligent automation or digital labor. And this is where AI has the ability to maybe do some manual work for us. I think all of us in HR are familiar with processes, maybe a promotion process as an example, where you've got to pull data from the learning systems and your performance management systems and maybe customer satisfaction systems to try to decide in a fact-based way who might be eligible for promotion or who you want to promote. If rather than me pulling all that information from reports and systems and collating it together, if AI can do that, that saves up a lot of time where I'm then freed up on what really matters. How do you make the decision? How do you have the conversation with somebody who's not getting promoted? Or how do you celebrate somebody who is? And I think that reduction of those manual repetitive tasks are really going to unlock a lot of value in our organizations.

Fuller: Well, certainly our research at Harvard Business School confirms several of the things you just said. It seems that generative AI really shifts both the amount of work that can get done and the quality of that work, although you do get some outlier results that relate to what some people call hallucination. I call it confabulation because it's not that it thinks it's seeing something that's not there, it's made up a story based on its best efforts to answer your question. But also that its real impact is to replace elements of work, not jobs in their entirety. And a lot of the early elements are routine. They're not creative. They don't require a lot of interaction. I think they are tasks that cause white collar jobs to be boring that are dissatisfiers. And let's take those hours off of people's diaries, make them more efficient and put those hours against learning or put those hours against development or better customer contact or more engagement with the community, or whatever else is called for. How are you approaching introducing AI to the workforce, familiarizing people with it? We've seen a lot of large companies that they're not investing in making their employee base comfortable with AI, investing in their just general understanding of it. How are you managing that process, Nickle?

LaMoreaux: As you say, for some of our job roles, our software developers, this is table stakes. But we are doing this across the organization, accounting departments, HR departments, our sellers. They are all getting this baseline training around artificial intelligence or intelligent automation. So to your point, it's helping them understand the technology, make it less fearful, what it can do, what it can't do, also be on the headlines around AI and what it can unlock. I think there's also an important point in our organizations that we all need to embrace, and that is experimentation. So it's one thing to do training and that's helpful, but do you allow employees to experiment with AI, with automation, so that again, they can get a firsthand experience of how it can be helpful, where it is not helpful or shouldn't be used right now because it just doesn't do that. Of course, you can put parameters around where you want experiments running and where you don't want experiments running. Two of the use cases I talked about can be tops down decisions, senior leaders can make them. Recommendations. Are you going to use AI to make recommendations in any of your processes? On something like a chatbot or a digital assistant, is your IT department going to launch a chatbot to answer IT questions? The last piece I talked about, which is this digital labor, digital assistant, that is not a tops down decision. That is every employee in your organization based on their job role having some ability to experiment with this about how it could make their job easier is another way. Now, this doesn't have to be every employee in your organization, but can you find pockets of early adopters that are willing to be those co-creators with you or willing to be those early experimenters with you so that you can then start to unlock for specific job roles.

Fuller: Well, Nickle, I'm sure our listeners would really benefit, do you have any rules of thumb or ways that you think people could know that their AI strategy is working? Are there two or three indications that you'd advise they track or you yourself are tracking?

LaMoreaux: Yeah, and I think this is an important question. Because as we talked about, we're in this age of experimentation. So we probably have lots of little pilots going. So how will you know when something's having an impact, whether you should continue it, whether you should stop it, and is this strategy working? At a very micro level, I would say there's a couple key factors. Are people using it? Are they engaging with it? So just usage rates. Adoption is always a key indicator. The second piece is, is it changing outcomes? And some of those outcomes could be, is it making the work go faster? It could be are employees more engaged or giving you higher scores on the experience because it feels more streamlined to them? So think about is it having that desired outcome? And I guess just the last macro piece I would say is, is it driving business results? I think there's lots of things you could do in AI that are AI or technology for the sake of technology. But as HR professionals, we've got to be thinking through, is it actually having an impact on your business? Are you getting better talent? Are you getting better returns? And that is the ultimate measure.

Fuller: Well, I think the measure for all HR policy should be, is it expressing itself in terms of improved competitiveness, retention, skills building? And in the old world of HR, it would've been, are we checking all the boxes in compliance, regulatorily, administratively? Is everyone doing exactly what we need? That world is behind us. And I think it's exciting to hear what you're accomplishing in IBM and also what a technology leader is doing to harness the technology it sells. Well, Nickle, thanks so much. Nickle LaMoreaux, CHRO of IBM, thanks so much for joining us on the Managing the Future of Work podcast. It's been a pleasure.

LaMoreaux: Wonderful to be here. Thank you.

Fuller: We hope you enjoy the Managing the Future of Work podcast. If you haven’t already, please subscribe and rate the show wherever you get your podcasts. You can find out more about the Managing the Future of Work Project at our website hbs.edu/managingthefutureofwork. While you’re there, sign up for our newsletter.