Das: For our next panel, I'd like to introduce Tina Mylon, Rachel Kay, and Chris Louie. As you guys make your way, I'll do some intros just so that we can keep things moving. Tina is the Chief Talent and Diversity Officer at Schneider Electric, where I believe currently you have an initiative upskilling at scale, called Upskilling at Scale. Rachel Kay is the Chief People Officer at Hearst, leading recruiting, diversity and inclusion, compensation, and talent planning. I'm sure quite a few other things across all of Hearst's businesses. And Chris is the Head of Talent Development at Thomson Reuters and also teaches, I believe, algorithmic responsibility for the Human Capital Management program at NYU.

Chris: That’s right. I work for Anna. 

Das: Wonderful. One of the things I love about a conference or a day like this, a summit like this, is that everybody comes from such different businesses and industries, but is talking about a topic that's shared. AI is going to impact different companies and different businesses in different ways, but it's going to impact all of us. So I'd love to just start by hearing, what is the impact in your industry and specific business, and then, if you're able to, what are some of the insights into the new skills and roles that you're finding you need? Round robin, maybe, Chris, if you want to start. 

Chris: Yeah, happy to start. So there are lots of analyses that are out there looking at different industries, looking at workflows and roles, and trying to estimate how many of those are either able to be augmented or automated with AI. If you take a look at legal, and if you take a look at tax and accounting, those are usually the ones that are all the way in the–whatever your two by two is, but– upper right, those happen to be our biggest businesses. Reuters is also part of that as a news business. And so, a very profound potential impact of a AI, both on our product set and then, because of that, you know, for our own employees, it's very easy for them to get their heads around the fact that AI is and can be existential, which honestly has made our job a bit easier. Thinking back to that change management panel a little bit ago, to help employees understand the importance of, you know, getting proficient in AI and continuing to develop skills as, you know, some skills that you may have rested your laurels on for the last several years–or even decades–may be either less relevant or less protected. So some pretty profound impacts, both on our external business and then that flows through to our internal. 

Das: Rachel?  

Rachel: Let me just do two examples. So Hearst actually is a portfolio of six very different industries. But two examples of things we're worried about–and I think we don't know what's going to happen– but, take magazines. Right now, magazines is largely a digital business. The bulk of our revenue in magazines comes from digital advertising. How do people find our magazines? They go online, they might search: what are the best air fryers? They click a link, it goes to the Good Housekeeping website. Now they're looking at our website, we're getting revenue from advertisers because you are seeing advertisements on the website and then if you go ahead and click on the link, to Amazon or whatever, to buy the air fryer, we're getting revenue from that. Those are our two biggest pieces of revenue. Now, when you go on to Google and search for one of the best air fryers, you're going to get a blurb at the top that synthesizes it for you. You're never going to come to our page. So we're facing, in magazines in particular, some really existential threats around two of our biggest revenue streams and how we are going to accommodate for that. 

At the same time, of course, all of the different–and you see these in the lawsuits–chatbots or LLMs are using our content to create the answer. So it is an interesting question around how we square that circle, because there's a lot that's unknown, and some of it will be legal, I'm sure, some of it will be other revenue streams, but we're really working that out. 

Another business that we have which is very different, Fitch Ratings. Fitch Ratings is a credit ratings agency. They compete with Moody's or Standard and Poor's, you think about what they do, those analysts, they're just combing through tons of information–much of it publicly available–to assess the risk of purchasing a bond from a particular company. That, theoretically one day, could all be done automatically. Why come to a ratings agency? Now, at the moment, you have to come into a rating agency because it's highly regulated and that's required. But in the future, will investors need that? Or will they be able to determine those things on their own? So I think there's some really big questions our businesses are struggling with, and so much of it is still unknown. 

Das: And that existential piece, I think, is an interesting one. Tina, you're in a very different industry. So what are you seeing at Schneider Electric and how are you feeling? What's been the impact on the company and some of the roles needed and skills? 

Tina: Can you guys hear me? Okay, just testing.

So hi, everyone. I work for Schneider Electric. So we are an almost 200-year old company headquartered in Paris, and it is half an industrial tech company, as we call it. So, basically in energy management, in terms of automation, digitalization, anything, a plant, a data center–that's a big part of our business–a factory is using to manage their energy efficiently.

So, our whole mantra is: energy access is a human right. And it's a distributed workforce: about 50,000 people in Asia, 50,000 in Europe, 50,000 in the Americas. Part of what we do is really try to make sure, at the end of the day, whether it's your home or whether it's a plant or whatnot, are you using energy in the most effective way? With electricity being one of those game-changing technologies we talked about in the morning as the most efficient vector. 

So, back to your question, part of what we're interested in, especially for the customers, is how do you create a more sustainable energy landscape? And what is the role of digital? And now, 10 times, 20 times, 100 times over, what AI–even generative AI–is doing? It’s a lot about managing the consumption of energy. So how can we be more efficient that way? It's a productivity gain. And we see a lot of interesting opportunities with AI, as well as generative AI. And then, also, what is the energy mix? So how do we shift also about more sustainable, like renewables, and how we do that. For me, in the talent and diversity space, that comes to also a very pragmatic question that all of us, I think, are grappling with is what are the most pragmatic use cases when it comes to making sure our workforce is equipped to, basically, serve our customers and also the broader sustainability goals that we have for society at large.

The last thing I'll quickly say, I mean, for us, one of the things I think that's most interesting to me–and we'll get into use cases–but a couple of the panelists and some of the speakers talked about inclusion and equity. And that, in the face of generative AI–wearing my DEI hat–is super fascinating. So really, in the advances of all this accelerated technology, how do we make sure no one is left behind, just like energy access? How do we make sure the data is as good data as we can do? We are all struggling with that, as you guys alluded to, especially around biases. And then how do you make sure everyone has access to that transition to a more AI-first world? 

Das: We have some great themes here, I think, to talk about. There's this existential threat, I think, that a lot of people are starting to feel. As a content producer myself. I know there's sometimes that feeling of: back against the wall, I need to change, or what's going to happen next? We have this idea of, how do you bring everybody along within that environment? And then you have the fact that  some businesses, it is existential, and others have huge opportunities. Within all of that change, and how fast everything is moving, where are you now with understanding, do you know what skills and jobs you kind of need in the future? Are you just figuring that out? Where, where are you with that question? 

Tina, you highlighted it as something that's being grappled with. 

Tina: Maybe I'll start and then have Chris and Rachel come in. For us, the answer is mixed, quite honestly. We are embarking, and a couple of folks I talked to, it's one of the things that's keeping me up at night–and Anna spoke eloquently about skills and what it is, what it's not. At Schneider Electric, we are embarking, and it's already been a year and a half in the journey, on a major revamp of our skills ambition. So, we're actually not even starting with technology. We're starting with the whole job and skills and career architecture. For those of you who have been in this space for a long time, you know how painful that can be.So we are redefining our job architecture. More outside in, more market data, and at the granular skills level. We've just engaged, in the last couple weeks, a technology partner to have the end-to-end user interface. And for us, it's really starting with the most critical skills, which I think will be familiar to all of us. Technical skills for our R&D area, especially around engineering, digital AI, though probably more even on software development for us, and certain human skills. This is where we're trying to codify in the system, in the user experience, where people are, what the gaps are, how do you have the pull through when it comes to truly upskilling people at scale.

And we are 150,000 people, like I said, so every year 15 percent are turning, through hiring, but the 85%, that's the mighty majority of folks that we want to very much be ambitious and also focused on how to support them in their upskilling. 

Rachel: Yeah, I don't think we know what the skills are we're going to need or the roles are going to be because so much of it still is unknown in terms of what our future business model looks like. You know, I think what I'm what I'm struck by right now is that the most important skill at the moment seems to be curiosity. And what I love about this and living through this moment is in some ways how egalitarian it is because nobody has the answers and so that means the answer can come from anyone. 

At Hearst, one of our, you know, new folk heroes is this guy named Mike McCarthy who was hired a year ago as a salesman in our Connecticut newspapers And he has, just like Jennifer was describing, when the ability to create your own GPTs launched, he was one of the first people just to dig in and start playing around. He created a whole raft of GPTs, you know, one is the skeptical buyer GPT, so a salesperson can go on and practice their pitch. Another is the deck creator GPT, he has like 10 different GPTs just for newspaper sales. And he started playing with him on his own, he got his team to use them, and all of a sudden, you know, other people across our newspapers businesses are picking them up. And now he has a new job: he's the head of AI for newspapers. So, you know, just thinking about, would you have picked a first year salesman from our Connecticut newspaper as the future leader of our newspaper's AI efforts? I never would have. If I was looking for the skills for that role, I wouldn't have thought, let me go find a guy who's really great at sales in Connecticut, but because we were open to it coming from anywhere and because he was curious, he was able to take that on. I just think it's really exciting and I think what I'm trying to check my own biases on is being too prescriptive on what the skills are that we need because I think we're still figuring it out. 

Chris: Rachel, I'd add on to your curiosity, important skill, I’d definitely agree with that. Two things I would add and put kind of at the same level. One is critical thinking. Because, you know, curiosity, you can ask the questions, you have the, the sort of, like, courage and the compulsion to ask the questions. There are so many different places where you can get answers back now, and as, as we know from, you know, kind of leveraging AI, a lot of those answers can be really, really wrong and hallucinated. I mentioned before, one of our biggest businesses is in the legal industry, and, you know, the cautionary tale that keeps going around are the couple lawyers who submitted, you know, case histories that were completely fabricated. And, you know, they got their slap on the wrist. You need that critical thinking, we all need that critical thinking in the world these days, but also especially in trying to leverage AI.

The other thing that I would add into that is, I guess I might call it imagination. Because I think the beauty of the promise, the potential of an AI solution is that it is giving us capabilities and skill sets, both as organizations and as individuals, that we might not have otherwise had or had access to. And as a result, you can imagine and invent both new avenues to go down in different ways of doing things. And, you know, to loop back to your point around trying to figure out, you know, how things should work and where the business is headed. I think that that is like, the first thing that you need to do in order to try to figure out what skills you ultimately need to develop. I don't think it really works the other way around. Where's your business going? How's it going to work, or how could it work? If you don't know, but you can have some scenarios and then, Okay, what skills do we need or will we need in order to be able to either explore or deliver on that promise? I think we all either are or should be kind of living in that in that space right now. We. We certainly are in Thomson Reuters. 

Then Rachel, the other thing I would, I would share is, you know, the example that you just gave of, you know, the salesperson that may not have been the stereotypical person to put on top of that project. You know, we did a similar thing with somebody in our news business as well, taking them out of their job, having them spend time with our Thomson Reuters labs to really understand the potential of what AI could do, and then unleashing them on the business. And, while again, it may not have been the stereotypical background, I think if you think about those different dimensions of curiosity, she definitely had that to even be up for this. Critical thinking, I do think that that's a hallmark of, you know, folks that have operated in the news industry, whether it was AI or just human beings telling them stuff, they have to really apply that filter and lens to what they're hearing to try to get at the truth. And then the imagination piece, I think, you know, what better person to dream up than somebody who's been living in one place but with the expertise of, Hey what was painful or what has been painful and what could be better? 

Das: Yeah. And it's interesting, as we're highlighting these skills, kind of the three big ones I heard were curiosity, critical thinking, imagination. And then even the word upscaling. And the big word that comes to mind for me is this idea of learning. And one of my favorite business quotes is from a guy named Arie de Geus, who said learning faster than the competition is the only sustainable competitive advantage, especially in these sort of fast-moving times.

So taking that kind of learning lens, how have you approached designing the learning programs necessary? I know you all have different initiatives around this. Talk a little about those initiatives. Like, how are you creating that space for learning so that you can learn faster than the competition, so you can adapt? And I think earlier somebody highlighted the innovator’s dilemma. Like, how do you get out of that? And what are you doing? 

Das: One of the things that we've done is actually to go right at what you just mentioned, of creating the space. So, you know, I don't know that this is completely novel. I think many companies have taken a stab at doing things like having a focused or dedicated or regular learning day or learning week or learning month. You know, and it sounds like, listen, we should always be learning, and therefore why do you need to just dedicate a day? Having said that, there's a lot of stuff we should be doing, you know? And it's just how do you demonstrate to your organization that, hey, this is really important, that it is a priority at the level of the other things that you are doing, and we are in a concerted way together going to block the time. 

Chris: So that's what we've done. Every quarter, we have a learning day. We instituted that in early 2023. Probably not coincidentally, the first learning day was dedicated to AI. So we had sessions on, you know, AI and LLM 101, AI in our products, AI in our internal operations, and then a workshop on AI in your job. We ran multiple versions of those sessions across time zones so people could get the benefit of live.

And, again, that demonstrated to people to make time for learning. And it also became a key pillar of our approach to getting our workforce kind of spun up on AI. We tried to go broad in educating. This, again, was an example of that education. Enabling, so we put effectively a privacy-protected version of ChatGPT in everybody's hands. And then experimenting, both with that solution and then through hackathons, etc., to just show people that not only was it okay to experiment, but that it was valued. 

And then the other piece of that AI learning was actually focused, right? So we had a bit of a SWAT team that would go in with individual organizations and with their leaders to help diagnose where the biggest business needs, and therefore potential AI use cases, might be, and figure out how technically, operationally, and procedurally, and from a human standpoint, we may help them to get over the hump to actually enable their workforce to realize those use cases.

So that's kind of the approach that we took. And they was specialized, function-specific learning that went along with those focused approaches. 

Tina: We're really trying to move away, and we're not really there yet, from learning hours. So coming from a very KPI, somewhat traditional learning environment around the world. When I joined the company eight years ago, it was like this obsession. Like, tracking every little thing, you had to mark it in your LMS. So we are similar. And I think we've borrowed your phrase as well, because our campaign is around creating time and space to learn. Now, the thing is, by freeing that and encouraging people to really upskill, at the same time, we are very focused on codifying that data.

So that's why, back to the skills transformation, having a system to do that. And sometimes it sounds a little harsh, but also saying: this is about us staying ahead of competition, growing the business. But you grow yourself, you grow the business. So that whole expectation is: it's your job. It's your job to upskill, stay relevant. The company has to do its part to provide the great user experience, the great content. But that mindset shift, we're trying to message more and more. It's not easy. Everyone's still used to: How many trainings do I take? Which ones are mandatory? Like, how many hours? And that broader shift is something we're trying to implement.

Rachel: So we're kind of a hot mess when it comes to, like, learning culture. Hearst is interesting. Whenever I go to these panels, I always feel, like, inferiority complex creep in, but, 

Tina: No, it's like collective commiseration. 

Rachel: We're incredibly decentralized, right? I mean, Hearst is incredibly decentralized. We have six very different businesses. And, from the center, we tend to provide carrots, not sticks, right? So here's stuff. If you want it, business, you're welcome to use it. And if not, that's okay. Our feelings aren't hurt. This is one area where our CEO kind of came down and said that we need people to be familiar with this.

So in terms of AI, I'm thinking about it as 101 and 201. 101 is mandatory, and 201 is optional. But in terms of the mandatory, we wanted to launch training to make sure everyone was familiar with the tools. Our tech team had invested a lot, and I know many businesses have, in creating proprietary versions of the OpenAI, the Claude, the Anthropic, all those different tools.

But we weren't seeing a lot of uptake. The curious, the early adopters were using it, but over 30% had never ever even tried it once, and the vast majority had gone and looked at it and never gone back. And so we wanted to force people to do at least something, but it had to be something of quality and something that felt relevant.

And so we invested in having live learning sessions. Everyone had to go to a live session, virtual live session. And it was customized based on your job function and your business. And so we had 17 different versions of the training, and we delivered the training over 100 times. We got 8,000 people to go through it.

The scores for the training were pretty high. They weren't out of the park, but, you know, my L&D leader kept getting very upset by that. I'm like, look, we dragged some people to this, so of course they’re not gonna say it was fantastic. But, in general, it got pretty good feedback, and it was customized, as I mentioned, based on what they do. So if you're a salesperson, you're going to go and see use cases for how to use this in sales. If you're in HR, the use cases were in HR. If you're in content and news creation, it's going to be use cases relevant to that. 

That group, by the way, was our lowest scoring, both in terms of value for time spent and also propensity to use genAI in the future. Anyone who's a content creator, based on what I was just saying earlier, are very skeptical that this is going to replace their jobs. They're very skeptical that this is going to downgrade the quality of what we do. And so, you know, seeing that, having them go through the training, getting the feedback, hearing what they're saying, has actually been really helpful in pulling together our future communications on that.

So, that was the 101. In terms of 201, we have genAI champions, we have our first Tech Academy, we have a TechNext Conference, where we have in external speakers like Ethan Mollick. So we're trying to keep it in the water now that people can keep getting smarter. But for us, the big unlock was making sure everyone at least had that first exposure, and then they could kind of decide where to take it from there.

Das: So we've come across this idea a little bit of measuring success, right? How do you measure success with upskilling, and how do you measure success on upskilling when you're trying to figure out still what those skills are? So I'd love to hear a little bit about, just if we could dig in a little bit more on what are the metrics and the measurements and how you're thinking about that.

Rachel: I mean, for us, and again, this is probably pretty basic. For us, right now, the measure is people's usage of the tools and active projects under discussion that use genAI. So, if you look before and after we launched that training program, we saw a 175% increase in the number of users using our internal ChatGPT. And we saw a 200% increase in the number of projects under active discussion for development. So, for us, that's how we're measuring it at the moment, right? Eventually, it'll be actual business results. But for right now, I think it's just people engaging, is our first horizon. 

Tina: I think, for us, just simply breaking down more maybe soft or human skills versus the technical skills and the digital skills. On the latter, maybe we're overengineering it, as Schneider typically does, but we are going hardcore in codification, certification, levels. What's helped me is it's allowed me to put more business case to also, frankly, outsource a little to go faster. So going to like a Coursera or a platform, where it’s fairly structured, you can cater it by certain domains. Because we have this habit of wanting to create and build because we think we know it all, and that takes up a lot of time as well. So on the latter part we are super structured about certain domains and job families, where you have to be, and some of it's very compliance driven.

Chris: We are having very active debates right now about this very question and the different metrics, including the activity and usage metrics that, Rachel, you just shared. And then trying to be more regimented about what do you need to actually have developed, and have you spent the time to develop it in these specific functions, et cetera.

Those are certainly some of the things that we have been measuring but that we are talking about, because we are trying to figure out: What are our top two or top three things, right? We all have like hundreds of things that, you know, metrics that we look at. One thing that we are talking about is trying to get past the activity, kind of to your point, and thinking about the outcomes. And one set of outcomes, from the employee or from the colleague standpoint that we've talked about, kind of gets at, I think, a core belief. Like, yeah, we talk about skills all the time. We, literally, we in this room talk about skills all the time. And I think to any individual employee who might be thinking about themselves, I don't think that skills is the thing. They can think about skills, but they think about skills as a means to an end. And I think that the end is often like near-term and long-term career growth, career opportunities. Having a job that I like or am excited about. Having things that I can go to that I might be even more excited about in the future or work toward. 

So in that vein, the engagement survey that we put out has, like everybody else's survey does, statements that we ask the degree to which you agree or disagree on it. And statements around, I believe I'm growing my career at Thomson Reuters, or I believe I'm developing the skills that I need to be successful, to be able to do my job or unlock future opportunities.

We're looking at those measures, and we're thinking about those measures very seriously. Those have improved, by the way, over the last couple of years. We want to keep that going. And so we're thinking about that from a colleague standpoint. From the business standpoint, I don't know how many of y'all talk about workforce planning all the time, or strategic workforce plans. We certainly do. And if you think about it theoretically, there's a case to be made of, okay, delivery against your really great crystal ball defined strategic workforce plans in terms of the skills that you will need. If we could somehow both get to those plans in a good way and then assess the degree to which you've developed those skills so that you're  hitting on your plan, you're delivering on your plan, which means that you have the talent that you need now and in the future to address business needs. That would be really great. You know, are you 90% of the way there? Are you 50% of the way there? A lot of the stuff I just said, we don't have the things solid, but, theoretically, if you did, that feels like something that could be really great and on target with delivering for the business.

Tina: It's not perfect, but we're doubling down so much on skills. And listening to you guys and the audience, if I'm going the wrong way, you guys need to tell me, and I need to pivot because I'm like, oh my god. But one quick thing that Anna said that really resonated with me, also. I totally agree, Chris, that skills isn't the end-all, be-all. It's the growth and the career evolution of employees. That's super important. When we make talent decisions, we actually have that sequence, meaning we do start with skills and qualifications to be able to do the job. And then we do two, and this is small scale, maybe that's the top 1,000 jobs, but we look at two factors. So you look at skills, and then we look at your preferences, a little bit more psychometric. And then we look at strategy. So you may have skills where you're a really strong communicator, and then, in your preferences, you're a huge introvert, like myself. And then you look at the strategy. Tina might be good at communications. She's a strong [communicator], but she's a huge introvert. Yet what does she do strategically? That's like, what is she actually doing about it? And that formula, we're testing it, and it works pretty well. That's back to the holistic, beyond skills, how do you really grow a talent and assess a talent? 

Das: So I know we're coming up at time, but I have to get one last question in here because you just teed it up so beautifully, which is: You know, we started the day talking about this great promise of AI to really personalize knowledge to each of us if it has the context it needs. And so, even though AI is what we're trying to adapt to, it's also this really powerful tool that can help us adapt. And I'd love to just hear how are you thinking about that, and how are you thinking about an AI or an AI coach as a tool for learning and that sort of upskilling and adaptability.

Rachel: I mean, whether it's a coaching tool or AI in general, AI is great at teaching things that it already knows. I mean, I see this personally with my kids, right? You can go on and generate quizzes for, you know, my son had to read the first 60 pages of Fahrenheit 451, ask six questions that would test to see if my son had actually read the first pages of Fahrenheit 451. Worked great, and he had. But I think it is great to reinforce things you already know, right? So I think it can be used either in this coaching context or even just, you know, I want to test my ability to articulate a philosophy for this or an approach to that. What are some questions I might get asked? Or what are some things I should be thinking about? So I think the tools are out there to use already that can hold yourself accountable in your own learning journey, whether it's customized for that purpose or not. A lot of them are just generally good at it. So, you know, I haven't thought yet about how to formally incorporate it, but something we tell people is: you know, once you've learned something, those same tools you just learned can now help you to retain that learning. And you should be thinking about how you do that on a regular basis. So that's just what I am thinking.

Chris: I'll add on to that of, I feel like with most problems, most business problems or situational problems, the answer is usually within the person. It's usually within you, and you just need something to help you kind of like pull that out. Most of the time, not all the time. Sometimes I have no idea, the answer is not in me. And I think an AI solution, a coaching solution can help. A great coach can help with that. I think a coaching solution can help with that. And then the other thing I think is, we don't scale, you know? What HR learning organization here is built to scale in anything that remotely looks like a one-to-one way, at least for most of the population, right? And so part of the beauty of this is, it can scale and it can personalize. 

Tina: For me, just to add, it is, to me, I think of AI, especially generative AI, as augmentation and acceleration. And nothing replaces the human touch. So high tech and high touch go well together, and the choices we make will have big implications.

I, myself, have been using Nadia for about half a year now, and we're piloting it in our organization. And it's going quite well. And I shared this with Parker, but, at the same time, my chief AI officer has been on my case going, “Pay attention to this, da da da da.” They were super nervous. But the way we position it, again, to augment, to accelerate, like a check in, as a leader, I myself have found that super useful. So, I'm positive, cautiously optimistic. 

Das: That is wonderful, and, you know, thank you again. I feel like we could take quite a few more questions on this, but, for the sake of time, I'm going to let everybody get to their break.

First though, I want to just give a huge round of applause for Tina, Rachel, and Chris. Thank you so much.