2025 Surprises and What to Look For in 2026

Speaker A: M. This podcast is brought to you by Mercer, an industry leader helping organizations create the digital advantage in the now of work. Thanks for being here and enjoy the episode. Hello, everyone, and welcome to the AI Evolution podcast. My name is Adriana Ocane, and, um, I'm your host for today. And I want to start this episode by saying thank you. Thank you for taking 20 minutes of your time to spend it with me and with us on this podcast. It's been a, uh, privilege to host these conversations over the past year, and today we're recording a special episode, not only, but our last episode of the year, but the last episode of the AI Evolution series. We will continue bringing you our perspectives and insights from a new channel, the New Shape of Work, uh, podcast. So for our final episode, I want to spend some time reflecting on what happened this year, what surprised us, and really the signals that leaders should be paying attention to as we move forward. So I couldn't think of a better person to have that conversation that Jason Averbook, for those of you who have been listening for a while. You know Jason well, he actually started this podcast a couple years ago, and he was kind enough to let me host it for the past year. So, Jason, welcome. Thank you for joining me today.

Speaker B: Adriana, thank you so much. And thank you for everything that you've done in hosting the podcast and, uh, and putting out a great set of information and content to people. We only get 20 minutes.

Speaker A: We can. We can take more time if we need to. I mean, it's like we're in charge.

Speaker B: We're in charge. Let's go.

Speaker A: We're in charge. It's the last one of the year. We can do whatever we want.

Speaker B: No, we'll try to go through. People don't have patience. I guess most people would do this on a treadmill, don't they? Or walking, probably.

Speaker A: Yes.

Speaker B: Like, people need to go longer than 20 minutes. I wonder how many people are sweating right now.

Speaker A: Maybe we should just get to the content.

Speaker B: Yeah, maybe we should. Yes, exactly.

Speaker A: Jason, for those of, um. Uh, for those who are just joining for the first time and who don't know you, can you share more about you and the role that you play at Mercer?

Speaker B: Yeah, for sure. Thanks for asking. My role at Mercer is I lead our digital transformation business on a global basis. And what that basically means is working with organizations around the world as they think about change, and that change could be how they operate, that change could be the technologies that they're using, that change could be AI. Today we're going to talk about AI, but change is not just tied to AI in the world today. You know, AI is the accelerant of a lot of change. But my focus and my focus really for the last three decades in this space have been helping organizations think about what change means to them, how to internalize it, and then how to externalize it when it comes to driving outcomes.

Speaker A: Uh, thank you, Jason. And I mean, you know, as I was preparing for this episode and I was reflecting on what has actually changed over the past year, it's like we moved from conversations around prompts and chats to more autonomous action. We've moved from standalone tools to really AI that is embedded directly into platforms and moving from experimentation to dealing with real risk and regulation and accountability and adoption at scale and building systems and rethinking how work gets done and how do we bring people along. So with that in mind, what are the two, three things that surprised you, uh, this year? Maybe something that you weren't expecting outside of the latest tools or models of the technology?

Speaker B: Yeah, thanks for asking that. I mean, 2025 has been a crazy year overall when it comes to the world of AI and the world of technology. It's kind of interesting when I look around my desk and all of a sudden I see like three computers going with three different monitors, and I'm like, why do I have all of this? And a lot of it is just because there's so much change. And trying to keep up with that change is, you know, it's a full time job, um, which is part of my job. Um, but there's just been a lot of it. And, you know, the change hasn't been with just with new technology models or hasn't been with new technology tools. It's been about impact, it's been about roi, it's been about jobs, it's been about the news, writing about the jobs, et cetera, et cetera, et cetera. So in preparation, Adriana, uh, I have created some notes. So if I look down once in a while, I wanted to make sure I said the things that were truly on my mind and said them in the way that I hope people can take back and use. So the first one to me was this concept that I titled Speed Exposed Fragility. That one is so interesting to me is because a lot of people jumped on the tools wagon, like, we gotta get people using Copilot, we gotta get people using Gemini, we gotta get people using ChatGPT, you know, and really they treated it in my mind like a feature, not like A flow. So they treated it as, hey, here's some tools, go try them, see what they do. Here's some tools. Let's measure adoption, like how many people logged in to them. But to me what that exposed was a fragility around AI on top of how we work existing. Just kind of exposed the fact that we need to reset and redesign how we work existing. I think that the adoption created like this weird chaos. You know, as someone who teaches at a university, people started saying, well, is that cheating? I know in my kids classes people have been saying, is that cheating? At work people have been saying, is that cheating? Like even to the point where Cs, C level executives have said to me like, we used AI to help create this. I'm like, so what? Like, um, they didn't used to tell me they used Google, or they didn't used to tell me they use an encyclopedia, but now they're telling me they used AI. So it kind of exposed this chaos, which is kind of like beginner's mindset, which is like, what does this actually mean? And I think that leaders overall kind of lost clarity as to what we were trying to do with AI or maybe never understood it.

Speaker A: Right.

Speaker B: Like it just hit us so hard that we pushed it out so fast that all of a sudden people said, wow, are we using this to automate our existing stuff? Like we've used technology in the past or we're using this to orchestrate a new way of working. And I truly think that leaders kind of lost their way in 2025 when it comes to this. So the first big surprise for me was this concept of people jumped at speed. And what it did is it kind of exposed a lot of other things like operating models, change, fear, things like that, governance, if we could, I hate to use that term, but safety, that people probably like, well, maybe we should have thought about that before we just jumped in.

Speaker A: That definitely resonates with me. And I think you're right that we treated it in probably in the way that we've always, uh, treated technology. Right. Really without thinking, uh, or really understanding it well enough to anticipate that the way that we do things fundamentally needs to change. Right. So it's not as simple as just making it available, uh, to people integrating into their existing workflows. I mean, yeah, you can do that, but that doesn't drive transformational change, which is what we're after. Right. And when you said we lost clarity, I was really thinking my mind went to, well, to me, expose the lack of clarity of what we are trying to achieve, right? Because yes, it can help us go faster, achieve a vision faster, it could help us enhance experience. But really the big opportunity is like how do we create value in ways that uh, were not possible before? Right? So that's a big opportunity that we have. But do you see organizations getting more clarity as we start the year? Or do you think that that is still, uh, a bit of a work in progress?

Speaker B: Uh, and part of it is we don't know, we don't know what the outcomes should be. We don't know what the possibilities are. We don't know. Adriana, you and I have been traveling the world doing art of the possible. Like what's possible? Whose job is it actually to educate people on that? Whose job is it to make sure that we're measuring success? Is that it's job? Is it the business's job? Is it the manager's job which is part of the business? Is it HR's job? Like that's still very ambiguous. And until we put in place a true North Star and until we put in place true measures of success as to what do we want ROI to be and to prove, I think that we're going to continue down this path and I think we're going to have, I would say in 2025, and I hate to use this analogy, we probably had a lot of um, near mid air collisions. Uh, I think in 2026 we're going to have a lot of mid air collisions if people don't put these governance and cross organizational north Stars and measures of success and visions in place.

Speaker A: Jason, I have a follow up question. Uh, you said who, whose job is it? I am observing more and more organizations appointing someone as a transformation to help connect the dots across all of the efforts. Right? Because I think at the beginning very much those organizations that were embracing AI made it a mandate to everyone, right? Everyone has a role to play, which is absolutely true. But when there's not one person accountable, it is hard for all of these efforts to stay aligned and coordinated. And my perspective is we need to start building a system, a new operating system, and that requires important changes in existing systems. And that can only happen when there's someone who maybe temporarily plays this role. Right. It might not be a job that exists permanently, but for at least in the short term it feels like there's a need for someone to be doing that. Do you have a perspective on that?

Speaker B: No, Adriana, I completely agree. And I think that if you think about the concept of artificial intelligence as a True orchestrator, not an automator. You actually need an orchestra leader. And if you think about that orchestra leader, that orchestra leader is standing in front of the orchestra hearing how all the sounds come together in a harmonious way. If you think about sitting next to the violins, all you hear are the violins. If you think about sitting next to the saxophone, and that's probably not in an orchestra, but sometimes all you hear are the saxophones. But when you're at the front, you hear the harmony. And I think that's what's so important about this, is that if we're really going to use AI as an orchestration tool, I need to bring together, there's my conducting, the orchestra. I need to bring together how all these instruments sound to create one meaningful output.

Speaker A: Absolutely. I love that, I love that analogy. Um, the song at the moment doesn't, doesn't sound quite right. Even if some of the instruments do in isolation.

Speaker B: Well, because no one really knows how loud or soft to play. You know what I mean? We might all be. Even if, even if we're all. I mean, we could go on with this analogy the whole time, you know, But I mean, even if we're all reading off the same sheet of music, which is rare, but let's say we all were, we still need someone to tell us how loud and how soft to play, how, um, emotional and how passive to be. You know, I've never been in an orchestra, so I'm sorry that I'm not using the right terms, but I hope it resonates.

Speaker A: Yeah, no, it makes complete sense. So speed exposed, fragility was the first one. What's the next one?

Speaker B: So the second one to me is really important is, you know, is AI a tops down or bottoms up type thing? And I think it's a middle goes up and down, if that makes any sense. And when I say the middle, I think it's line managers. I think line managers and middle managers are the key. So when I wrote this in preparation of this, I said management got compressed, not eliminated. And what I mean by compressed is we pushed a lot of tools on people from the top. We had a lot of employees using things like their phone and real outside of work using tools. And managers started to say like, holy cow, we're not doing enough at the enterprise level. And they're hearing that from their employees, yet they're getting these things from corporate saying, we're working as fast as we can. So managers got compressed. And to me, managers are the number one indicator of the ability for AI to Drive and to get embodied in organizations. And you've heard me speak before, Adriana, where, you know, I try to use embodiment instead of adoption. Okay. Which is how do I make sure that I'm being in an AI driven world, not just using the tools. Managers need to create a safe space, managers need to create trust. Managers need to lead their meetings and making sure that AI is part of those meetings. Everything from how notes are taken to how groups are having conversations to saying it's acceptable to fail using these tools. And maybe we go back to the way we were doing things before and to drive first principles thinking from the ground, not from the top. So I think management in 2025 got compressed and I think it's going to need to play a much middle management, a much bigger role in championing these tools, in championing the employees and making sure that the once a week training course pushed out by corporate or the once a month training course or once a quarter training course pushed out by corporate, that's not enough. This is a much bigger change. And I need to create a stage for my employees that they can feel safe, that they can experiment and they can show off what they've done. So for me, that's the second thing was a surprise to me. And when I say it's a surprise, you know, I don't, I definitely don't have all the answers. Um, you know, As I entered 2025, I did not think that this was going to be such a heavy lift on middle managers. And I'm really starting to think it's going to be a heavier lift than what any of us thought going forward. Because those are the people that the employees trust most. They don't trust hr. Sorry. They don't trust it. Sorry. They trust the person that's judging, if I could use that word, and evaluating their work.

Speaker A: I mean, Jason and I agree. And this is making me think of COVID Right. Like, I mean, managers have had a ton of pressure and this is, I mean, if you. We look at the trend, I mean, how many years has been of a different type of pressure and in a group where there's not enough focus. Right. Uh, there's a lot of focus on top leadership and other areas. Right. Like there's been a lot of conversation on entry level positions, but I don't hear enough about what are we doing to equip managers. So.

Speaker B: And not if I could. Sorry to interrupt. It's not just equip them with the AI information, but it's to equip them you know, their employees have this, you know, fear of becoming obsolete. So how do I help employees realize that their jobs are going to change no matter what? I'm open and honest about that and I, as a manager, I'm going to help shepherd you along through that change. Like, we have left managers high and dry in my personal opinion, when it comes to that. Like, we have high level messaging, we have employees trying tools, and in the middle, managers are feeling the heat from engagement, from we don't have good tools. So I'm going to look for another job tool. I'm going to go work for a small company instead of a big company where there's not governance, et cetera, et cetera, et cetera. Managers are right in this middle of this and I think we missed them in 2025.

Speaker A: Yeah, no, you're right. And just to use your, your words, right, like, how do we help them embody that change so that they can help the organization transition this, but help their teams who have a lot of fear. Still a lot of questions. And there's, um, this, this change, integrating AI in your day to day, it's not something that you get good at in a day. It is trial and error. Right. It is a creation. It means really creating new habits. Right. And as we know, it takes time to create new habits. And this one requires really changing, uh, a whole lot of habits where there's so many tools, our environment is changing. So as we're learning new habits, also our environment is changing. So I mean, I couldn't agree more with your point. So that's number two. What's number three?

Speaker B: You know, number three. I'm going to start with a personal question for you, if it's okay.

Speaker A: Yes, please.

Speaker B: So have you ever worked out a new muscle or a muscle that you haven't worked out for a long time and it felt strained or it felt like, wow, I haven't done that for a long time. Whether it be playing a new sport, lifting a new weight, doing a new stretch, whatever.

Speaker A: It is very timely question because this week I took time off, um, and I started working out again. So, yes, a lot of, a lot of muscles, uh, that I forgot I had are hurting at the moment.

Speaker B: Yeah. So, uh, wow. I asked you the question. I didn't even know that that was going to be what you were going to say. But to me, a new muscle needed to be exercised and formed and strengthened in 2025 that I didn't think about enough. It was like I said, some of these are surprises to me. Adriana someone else might be saying, duh, you're so stupid. Uh, but to me, that muscle is judgment. And you know, we've lived in a world for 30, 40, 50 years where everything has been what we call deterministic. Like, you know, if this Excel spreadsheet gives me this answer, guess what? That's the answer. If this word document tells me that okane has an apostrophe after the O, guess what? It's right. I've never had to actually think like, I mean, I probably should have, but I probably have trusted tools and trusted processes so much that I've become radical. You know, not rad like cool dad rad, but rad like repeatable, auditable, and documented. And one of the things that AI demands is it demands us to flex the muscle of judgment. Okay? So I mean, something is basic. So when a machine gives me some information, is it right or wrong? Like that's very, very basic all the way to, you know, hey, I just created a graphic that this machine. A great image that this machine helped me with. But you know what, it's insensitive to people in Latin America because of A, B and C. Or you know, the machine told me to say Happy Holidays to a global audience, but guess what? Not everyone celebrates a holiday. So this is where I believe that we didn't think about human at the helm. I think we went into 2025 thinking about human in the loop. I think we're leaving 2025 saying, there's a human at the helm here, and that human has judgment and that human is still needed to say, hey, I've got amazing amounts of data. But guess what? I as a human have an amazing brain and I have an amazing heart. I might not be as good as crunching data as this machine. I might not be as good as spitting out words as this machine. But guess what? I know my audience better than this machine. I know how they're feeling better than this machine. I know what resonates and what doesn't resonate better than this machine. And there's a bunch of people listening to this that are saying baloney, like the machine knows better. I don't buy it. Like, mhm, this is a, this concept of a 5050 economy that we're in right now with humans playing 50% of the role and machines playing 50% of the role. What that requires and begs is us to leverage that judgment muscle. And for me, a big surprise was how much of that judgment muscle was needed. A and B, how many people just do their work like Repeatable, auditable, and documented. And don't use any judgment. And I think that's a muscle.

Speaker A: Yeah.

Speaker B: If I could say I want everyone to go to the gym to prepare for 2026, that judgment muscle, to me, is going to be so crucial.

Speaker A: Yeah. And, Jason, what you're making me think is, I mean, personally, this year has been a year of a lot of reflection. Right. When I think about what organizations we're all going through, it has really forced us to think, okay, let's pause. And again, where am I going? What are we trying to achieve as organizations? What is it that makes me special, that makes me unique? Right. That adds value. And I think that's a question that we need to be thinking in. Uh, a reflection that is perfect. So appropriate for this time as we close the year. Right. We were speaking earlier. We were talking about the lack of clarity. Right. And, yes, we don't know what's coming tomorrow. We don't know where the technology is going, how it will evolve, how rapidly. But if we have clarity of what we, as individuals, as organizations stand for, what is our purpose? What are our values? What is the value that you want to bring to the world? If you have that clear, it's a lot easier to navigate these decisions of investments. Where should I focus my attention? What are the changes that I need to be making in my portfolio of services or products? Right. So I think that that's something that, for me, didn't. I didn't realize it was. It would require such a deep reflection to kind of like, to the core, what. What are humans good for?

Speaker B: Yeah.

Speaker A: And then what is our role as we work with the machines?

Speaker B: You know, And Adriana, really, this is why I get so mad when people say AI is making us dumber. Like, literally. I have a visceral reaction to that because, like, I feel like I've had to work my rear off in 2025 because I'm pushing on these muscles that I haven't pushed on before. You know, AI's maybe making me dumber and doing dumb work, but AI is making me a lot smarter in pushing on this judgment muscle than I've ever felt before. You know, uh, we can put a link into the show Notes. My personal podcast. I just did one this morning where I. The reason that, you know, we don't. We're kind of struggling with navigation and what's next is because, as I said, uh, on this podcast, the map has changed. You know, and you talked. You used the word navigate a second ago. It's really hard to navigate on an old map. We don't know where the roads are, we don't know where the roads lead. We don't even know where we're going. Uh, and that's what a lot of us have been kind of plopped into the middle of this map saying, where's my career going? Where are jobs going? Where is education going? You know, I've said it from the beginning of the year, so this isn't a surprise to me, that this technology is going to cause a massive m. Massive reset. And it has, and it's led us to the point of the end of 2025 where people are like, what's the new map? And I think that's going into 2026 going to be our biggest opportunity is to create that new map.

Speaker A: Right? And you said opportunity because that's what it is. It's up to us, right? We have a choice to make. And yes, we can approach it and we're thinking less, we're doing less, um, analysis if we choose to do it that way, or we can use it to push us to do new things, more strategic things, right? Like, and that's what you and I get excited about, right? Like for the opportunity to do more impactful work, more of the things that we love doing that we're good at, and leaving the more transactional stuff AI to deal with.

Speaker B: One other thing that I think is really important is, uh, you know, how that repeatable, auditable work, you know, it might feel comfortable. One of the things that I found and I would encourage people to try is it also makes you tired. Like, I find at the end of the day, now that I'm focusing on judgment, learning and rethinking ways of work and what it means to the world, I have more energy at the end. Like, I don't want to stop. At the end of the day, I don't want to stop. Whereas doing that repeatable, auditable work, it's just a suck. It's just like, uh, oh, my God, seriously, I have to open another spreadsheet, you know, et cetera, et cetera, et cetera. So I think that not only are we going to be more productive, I think we're going to get a whole new set of energies that come from leveraging these new muscles that have atrophied for so long. I'm not saying that's your muscle, but I'm saying in general, I think that we've, uh, we have a lot of muscles that we haven't probably used since we were little kids.

Speaker A: Absolutely. No, I love that. I love that. And that requires curiosity. Right. And willingness to be uncomfortable, but there's a ton to gain.

Speaker B: That's exactly how your kids are, right? Yeah, that's exactly that mindset. And I think that as we get older, that atrophies in us because of what work has been like. Uh, and I think that, like I said, that's our opportunity in 26 to change it.

Speaker A: Well, thank you, Jason, for such a thoughtful conversation, for preparing. I really appreciate that. No, um, seriously, thank you for joining me, for sharing your perspective in this special, ah, episode. Um, I think we still have time for a final question. My opinion is that leaders at all levels at the very top of organizations need to be engaging with the AI tools directly and continue to be curious of what's out there. And I know you are constantly doing that, so I'm super curious to hear what you're experimenting with.

Speaker B: Yeah. Without getting too deep into the tools, I think 2026 is going to be a year, and let's just use the. I mean, this is actually a term, but let's just tie it back into the orchestra of, uh, what's called composable software, where we're going to be composing the tools that we need in order to do our jobs in the most efficient, effective and optimal ways. And to me, like, if I was pre recording this for 2026 and saying at the end of 2026, what was your biggest surprise it would be how much composable software has changed our jobs. M. Now, really quickly let me explain composable software. Right now, composable software is called Vibe coding. And Vibe coding gets this bad name. Well, excuse me, it can get a bad name because it's like, oh, it doesn't have any enterprise security. It doesn't have so and so. It doesn't have any so and so, which I totally agree with. But the speed at which these tools are moving, whether this is going to affect your personal life, which then allows you to show up at work in a better way, or whether you start building composable tools for your teams to interact. You know, example of that would be something like OpenAI's ChatGPT teams capability, which I wrote about a few weeks ago that I said I think is the end of email as we know it. You know, whether that be a vibe coding tool like Mocha, which is an amazing vibe coding tool, you know, very similar to a lovable or a replit, for example, that I had Mocha build me an exercise app Yesterday for the 12 days left in the rest of this year based on the things that I wanted to focus on. I composed software that works for me. So instead of me trying to figure out how to work with a piece of software that wasn't built for me, it was built for the masses. I composed tools for me. And for me, I think anything that we can do as humans to start to say, how can we compose? And then how do we start to think about the agents that are working with us? I think that's where 2026 is. It's all about us taking control, us composing and us defining what the new world of work looks like.

Speaker A: Great way to close. And I do have follow up questions, but I will save them because otherwise this is going to turn into a much longer episode. Uh, we will cover them offline then. Jason, uh, thank you again so much for, for your time, for your trust in hosting this, this podcast over the past year.

Speaker B: And I'm excited to see us on the first channel. Yes, that's what I was going to say.

Speaker A: Yeah, that's what I was going to say. You will find me, you will find Jason, Tara Cooper and many of our subject matter experts in the new shape of work. So make sure you follow us, make sure you tune in at, uh, the first episode of the year. And to close, a big thank you to all of our listeners for joining us. Thank you for listening and we will see you in the next episode of the New Shape of Work podcast. Thank you, Jason.

Speaker B: Thanks, Adriana. Happy holidays.

Speaker A: If you have questions or are looking for more information, reach out at the link in our bio. We can't wait to connect with you.