LH122 The Changing Shape of Work with Josh Bersin (FULL VIDEO)

L&D is behind. I mean, if you think about HR, recruiting people have been using AI for a decade or longer. Employee services call centers have been using AI for a long time. L&D, this is new. We haven't had this level of automation in L&D ever. And I think it's the most exciting area of all.

This podcast is supported by Synthesia. Training videos shouldn't take weeks to make. Synthesia lets you turn text into professional multilingual training videos in minutes. It's fast, scalable, and so simple anyone can use it. You can write an email, you can make a video.

Hey there, my name is Alex and I'll be guiding you through this training session.

70% of Fortune 100 companies use Synthesia to modernize onboarding, simplify compliance, and scale global training. Podcast listeners get an exclusive 30-day free trial. Go to synthesia.io/greatminds. Don't miss it. This offer won't last. Welcome to The Learning Hack, a podcast about the people and technologies that are creating the future of learning. I'm John Helmer.

Now guess what? Learning is cool. Learning is cool. Learning is cool. Learning is cool. I'm learning. Learning is fun. Knowledge is power. Knowledge is key.

Learning. AI crossed a threshold in 2025, moving from experimentation to reality. Now jobs are being transformed, skills velocity is everything, and L&D faces a structural reinvention.

The learning management architecture of an LMS with courses is about to disappear. It's not going to disappear in a year, but it's going to disappear.

For the first podcast of the new season, I wanted to set our horizons for the coming year. My guest this time is so famous in our industry for his ability to do that, he needs no introduction, but we're going to give him one anyway.

Hack Facts.

Josh Bersin is one of the most influential analysts in the global HR, learning, and work tech space. Founder of the Josh Bersin Company, he has spent decades advising CEOs, CHROs, and learning leaders on talent, skills, leadership, and digital transformation. His research has shaped how organisations think about learning culture, skills velocity, and more recently, the impact of AI on work.

We tend to treat analysts like Delphic oracles, but in reality, they can no more foretell the future than you or I. What they do bring, however, inhabiting as they do a data-rich environment, close to the cutting edge of change that organisations and their workforces are experiencing. Is a fix on the way our world is changing and the key decisions you might need to make in the coming year.

It's a great pleasure to have you on the podcast again. Welcome back to The Learning Hack.

Thank you, John. It's great to be here. Lots to talk about, as always.

2025 was a year of great change, and I don't think anyone expects 2026 to be any calmer. Organizations will have been gearing up for fresh challenges of a new year. What fresh hell is this? When we last spoke, you were already seeing major shifts driven by AI, particularly, and new models of work. Looking back over recent months before we turn to look forward, what have been the biggest moments or inflection points that you've noticed, or movements for that matter?

Well, I think 2025 was a year where AI went from an experiment to reality. And I think 2026 is going to take AI to an enterprise solution and enterprise application area. And I think in L&D in particular, there's going to be an acceleration of transformation taking place, which we can talk about, you know, in detail. And I don't find any companies that are not using AI at some point now, certainly in their individual productivity initiatives and emails and Copilots and Galileo, things like that. What hasn't happened yet is the big AI transformations of all the business processes in the company, including HR, including L&D. But it's starting and it's picking up speed and it's affecting people's jobs and roles. And it's surprising to many people how the job they love is about to be pulled out from underneath them and transformed. And that's just the way the world goes. We all have to reinvent ourselves on a pretty regular basis, and it's going to be a lot of that next year.

So let's drill into that a bit. You've described the emergence of the super worker as a defining feature of the AI era. Later in the year, The Supermanager. Now this weekend I was listening to a really interesting podcast, the Exponential podcast with Azi Mazar, who was interviewing Ben Zweig of Revelio, who's very close to all the workforce data. And he said what he thinks he sees at the moment is that there's a big lack of entry-level jobs. I mean, that's been noted by a lot of people and young people are noticing it as well. They're leaving college there's a lot less jobs for them. He thinks that this is more about the expectation from the C-suite of how AI will change work rather than any kind of big upsurge in AI-related productivity. But he also thinks that long-term, those entry-level jobs will kind of disappear and everybody's job is going to become more managerial. So a lot of super managers, maybe not so many super workers, and it'll all be about orchestrating different workflows. That'll be the important skill that people have rather than doing the stuff. I mean, obviously you still need people to do things like care homes, take blood from humans and so on. Robotics isn't moving that quickly, but a large number of clerical what we think of as clerical jobs will become managerial. Could you react to that for us and give us your take on that?

Yeah, I'm not a big fan of those kinds of forecasts. I don't think that's exactly what's happening. I think every job and every business role is being transformed in different ways, but they're all being transformed and there will still be entry-level roles. There will still be entry-level jobs. Companies hire entry-level people as a talent pipeline. You can't, you know, you don't wanna pay the salary of a manager when you don't need a manager. You want somebody new. Plus younger people tend to have more creative ideas. They're more up to date on new technology. And there will be lots of tactical jobs still needed to manage AI, to curate AI, to monitor AI, to train AI. So what we currently consider, like I'm old enough that I remember a steno pool. When I worked at IBM, we had typists typing on, Selectric typewriters, and you would walk into the steno pool room with a piece of paper that you had written by hand, and they would type it for you into a letter. Those jobs obviously disappeared when we had word processors and then PCs, but those people are probably either executives' assistants or they're working in sales or customer service. You know, if all the customer service jobs are automated by AI, the entry, quote unquote, entry-level customer service job is now a more consultative job. I was just on the phone with the CHRO of ServiceNow. They have used AI to automate their employee service, HR service group, and they went from a 400:1 ratio to an 800:1 ratio. So essentially they needed half as many people. But she said, what we're doing with the people that maybe don't have to do those kinds of jobs is now they're spending more time on more important cases and more important situations. And I don't know why people keep writing articles about entry-level jobs going away. There will always be— I don't want to— in our company, we hire people right out of college and right out of graduate school because they have incredible energy and new ideas that we don't have as senior people, and they don't make as much money, so it's not as costly. I don't think entry-level jobs are going away, and I don't think everybody can be a manager because somebody coming outta school doesn't have the business experience. They don't understand the P&L, they don't understand the operational issues. It takes time to become good at business. You don't learn it in school, you learn it by doing it. So, you know, I think there will be, you know, maybe more sophisticated entry-level jobs that, the, in today's definition, but you know, 2 or 3 years from now they'll still be called entry level.

So what's happening in the job market?

Well, what's definitely happening is virtually every job now has a new version of itself of a super, super worker where you're using AI. So if you're a salesperson or a marketing person, you're using AI to identify leads or generate emails or qualify customers or better write proposals or find new pricing or get training. If you're an engineer, you're using AI to write code. If you're an HR person, you're using AI to assess skills or identify the right candidate. If you're a project manager or, you know, a leader of some kind, you're using AI to do performance assessment and to evaluate project status. I mean, all of the jobs that we've created over the years are becoming enhanced by the superpowers of AI. And then over time, where I think that this is gonna go is the vertical job families are gonna overlap. The job families we have in, you know, sales, marketing, product, engineering, research, finance, HR, legal, IT, et cetera, those are industrial-designed job families with sort of vertical career paths, assuming that we would, you know, sort of stitch all those vertical things together into the horizontal workflows of our company. One of the unique things that AI does is it actually creates cross-functional solutions. So if you think about, for example, I like to use the self-driving car as a good example. A self-driving car is not an automated steering wheel, automated brakes, automated windshield wipers. It's something else. It's a cross-domain solution that takes you from point A to point B and uses all of those vertical things in a much more integrated way. So, you know, 3, 4, or 5 years from now, we may not have these vertical kind of job career paths that we do today. We'll have more horizontal ones. It'll take a while because these agents aren't all built out yet, but And I think everybody has the opportunity to participate in this. The steno pool folks that I worked with at IBM in the '80s, they were very smart people. I'm sure they found other things to do. IBM probably redeployed them over time. So, you know, the main thing is to not be afraid of this stuff and to learn about it and use it and, you know, take it, you know, as one of the CHROs put it to me the other day, think about AI as an opportunity to eliminate the mind-numbing, boring, tedious stuff that you didn't want to do anyway and to make your role better. And I think most of us have that opportunity.

Yeah, that's got an optimistic take. I think from what you're saying there though, just to try and put a kind of darker spin on that, is it dangerous to be a specialist in in the new AI-enabled job market? Because you've talked about structured jobs versus unstructured jobs. And some people are saying that it's different to have kind of extreme focus on a narrow area of domain expertise because that's what AI does so well now.

Well, you know, I don't know. I mean, a lot of people like to pontificate about the future like this. I don't tend to do that. There will always be specialist jobs. Yeah. If you take about, think about in HR, this job called people analytics. When I got involved in analytics in the early 2000s, it was pretty stupid. It was downloading data from a bunch of HR systems, sticking 'em in Tableau or Excel and trying to do some correlations. You know, that all happens automatically now., but there's still a need for someone to look at the data. Now we could think of it as a people scientist. Maybe we need to give it a new title. And so this job that used to be a very mundane kind of operational kind of tactical job of collecting data, cleaning it up, and trying to make sense of it is now much more strategic, but it's still a people analytics kind of a job. And it's not going to be any less specialized than it was before. Because now you're gonna see all sorts of things about the data that nobody else sees because you're a specialist. And if the AI gets smart enough to see what you see, you're gonna look for the next thing. So see, the thing that I think's unique about the human brain that's so different from AI is we're value-creating animals. So if the robot or the AI, whatever it is, takes away some stuff that we were doing before, we don't just sit around and eat bonbons all day and watch TV. We think about, oh, well, now that the robot's doing that, I'm going to work on this, which is a higher level or more sophisticated solution of some kind for my company. Now that the more mundane thing disappeared. So I think there will still be specialized jobs, data management, security. I mean, sales is a specialized role. Marketing has still got a lot of specialization within it, but there will be more cross-domain specializations where you don't specialize in interview scheduling. You specialize in assessment and selection because interview scheduling is done by the machine and you don't need a human to do that. So the The name or the title of the specialization may change, but there still will be specializations for sure.

Yeah. A total fail to try and bring you into my doomer narratives.

So I think we'll move on. I'm not a doomer. I'm sorry. I've been around long enough that I've seen these kind, this thing does, the AI doesn't intimidate me. I was around before we had PCs, you know? So I've kind of seen this.

Yeah. It's interesting what you say about sex, 'cause my wife, trained as a secretary, but right from the, it's in the '80s, right from the off she became a PA. PAs are still around. We don't have secretaries.

Yeah, I mean, think about, I have an assistant who helps me with my stuff, my travel and my schedule and everything. She has a very complex job and she uses all the AI and tools she can, but it's still a human job. So it hasn't moved as far as we may think yet.

Yeah, in a medium-sized company, very often the CEO's PA is the one who runs the company. And it's worth remembering that we used to have people who were called computers who did adding up before we had machines. Anyway. So skills remains a perennial issue, or skills remain a perennial issue, can't quite work out which it is. But you've been talking recently about skills velocity, the speed at which organizations learn and adapt. What examples have you seen of companies getting this right, and how do you see it playing out in 2026 for L&D teams and HR teams?

Well, I think, I think we took a little bit of a harebrained tour here into skills taxonomies, which I wasn't a fan of in the beginning, where everybody thought if they built a taxonomy, their company was suddenly going to be 10 times more productive. The AI can now do that for you. But this idea of skills-velocity is, to me, it's actually fairly simple. You're a professional or worker in some role and there's some new thing, a new tech, a new customer situation, a new product, a new pricing model, a new contract, a legal change. How quickly can you as an individual adapt to flourish with that new thing. You can call that a skill. You can call that a capability. And what allows people to adapt and learn quickly is very much cultural. Does the company give you time? Does the company give you resources? Does the company give you tools? Does your manager help you or support you in your learning? And the companies that are flexible and create opportunities for reskilling or upskilling tend to be much higher-performing companies. And we've done a lot of studies on this the last 10 years, and these more dynamic companies are constantly reskilling themselves at an individual level. Now, you know, if I'm a typical working person and I have kids at home or, you know, other issues, I don't have all day to reskill myself. I'm going to do my job and then I'm going to go home. So we have to create a work experience where learning and skilling is not a thing that you have to take time to do on Sunday evening, but it's actually something you do as part of your job. And these companies that have high levels of skills velocity talk about it a lot. They have learning, like they have meetings where they talk about new discoveries. They give people access to information. They have great learning and development environments. Expertise is valued and shared. There's storytelling. We call this learning culture. I wrote a big report on learning culture long, long ago. I think it was around 2006, 2007. But it's still true that one of the biggest drivers of skills velocity is valuing learning, making it available to people, and honoring the need for people to learn. You know, one of my bosses years ago was a guy who used to work in the military, and he used to make this joke all the time that in the military you're either training or fighting, and when you're not fighting, you're training. Well, I actually think that's kind of the way business is. If you're not in the middle of something urgent, you should be skilling yourself and learning something to make yourself better at your job. And that's very cultural. Some companies facilitate that, some companies don't. Some companies will say, they will label you and say, you are not skilled at this, therefore we're going to fire you or we're not going to promote you. When in reality, I think everybody has the capability to learn under the right conditions. So this is what CLOs are supposed to be doing is understanding this culture and trying to drive this culture through the HR function, but also through the management function too, 'cause managers have to be involved in this.

Yeah, a kind of perennial complaint that we get from L&D people is they just don't, employees just don't have enough time. To do that kind of learning.

Well, that's baloney. I mean, that's, well, there's two reasons that probably comes up. Number one, maybe you've created a bunch of courses that don't seem relevant and they aren't as aligned with the needs of the group you're serving as you think they are. The other is, like I said earlier, the culture of learning. If you're being asked or hired by a business person to help them with a learning problem, you should take it upon yourself to go back to that business person and say, if I develop this program or solution for you, are you going to make sure people use it? And you're going to make sure people take advantage of it? And if they delegate it to you and they think it's your problem to train them, well, that's very much mistaken. This has to be a two-way street. So I think L&D people have to be not just great creators and innovators and service delivery people, but they have to push back and educate managers on their role in the learning process. Sometimes that's easy, sometimes that's not. But I think somebody saying, I don't have time, is more a symptom of a problem in the process than it is a matter of time. Do you remember when Howard Schultz was at Starbucks a long, long time ago? One of these days, there was a big DEI thing going on at the time. He said, okay, we're going to do DEI training for everybody in Starbucks. We're shutting down Starbucks for a day. The whole company, they shut down the whole company for a day to make time for training. I mean, you know, you don't have to be that radical about it, but there has to be an accommodation in the whole culture of the company that taking time to become better at your job is part of our company.

If you're looking to connect with the top minds in learning and talent, there's no better place than an iVentiv event. Since 2009, iVentiv has hosted over 300 events for global heads of L&D and talent. I've attended multiple iVentiv events myself and found them to be invaluable. Less about sitting and listening and more about real impactful conversations. It's an unmatched opportunity to collaborate with leading chief learning officers on the challenges they're tackling today. Join the conversation. Visit iventiv.com/learninghack.

That's iventiv.com/learninghack. It can come from another direction too, the objections, in financial, pharma, a lot of sectors, there's an awful lot of compliance learning to do. People I know who in financial companies have to do masses of compliance courses a year. And L&D can say, look, we want people to do this, that, and the other kind of development activities. And the line managers say, well, you know, you're already taking up all their time with these kind of hundreds of compliance courses a year, and now you want to kind of put an extra load on it. The companies that you're seeing that are really doing it right is, I mean, to make time for learning, very often it's a matter of you don't do something else. Do they kind of make those kind of sacrifices, or are you saying there's never? A situation where you have to make a hard choice.

Well, John, I hope the dynamic is not the way you described it because the way the dynamic should work, and I think L&D people know this, is the L&D person should be a consultant or an advisor to help the manager improve performance. And they should be looked upon as a valued resource, not somebody that's throwing stuff over the channel and saying, take this or else. Now, compliance— nobody particularly likes compliance training, and anything you can do to make it faster and more efficient is good. But, you know, there's a reason there's compliance training, because it actually does remind people of things that they may have forgotten that they need to know, especially if it has to do with safety or government regulations. But the, you know, the ideal role or model or archetype of an L&D professional is as a consultant, not somebody who's like a drug dealer pushing drugs and saying, hey, take my drugs, they'll be good for you. And if you're closely, and this is why L&D has to be more embedded into the business, by the way, which is one of the things going on with AI, is the more embedded you are in the business, you see the day-to-day issues and you can volunteer to support whatever the issue may be, with the support of managers, and you're considered to be a valued resource, not somebody who's wasting people's time forcing them to take courses that they don't understand they need to take.

A lot of your research shows that the role of L&D is changing from content creation and delivery to acting as architects of skills, performance, and AI-enabled workflows. What, in your view, are the biggest shifts in the L&D function right now, and where is that evolution taking it next?

Well, there's 2 or 3 massive things going on. The first is the learning management architecture of an LMS with courses is about to disappear. It's not gonna disappear in a year, but it's gonna disappear. And we're gonna be building content dynamically from AI that is generated largely by the machine with support by humans. And then the employees are going to consume the learning in a chat experience or a super tutor. We call it the super tutor in Galileo, where they'll be able to ask the learning platform a question. And by the way, it won't even be a learning platform. It'll just be a chatbot. And then the chatbot or the learning platform will give them as much information as they need, and it will remember who they are and it will prompt them or coach them over time. So this whole idea of publishing courses in the LMS using SCORM, which is a very primitive technology to measure it, and then promoting the courses is going to disappear. Now it's not going to disappear overnight, but that's the big change. So instructional design is about to change radically. Content development and content publishing is going to be much, much faster and much, much more dynamic. Building videos, building graphics, translating into multiple languages, all those kinds of things are going to be done by AI. You've seen what Sora can do in the new Gemini models. I mean, this stuff is 100 times faster than it was before and 100 times more high fidelity. Assessment, you're going to have to think more about your job as listening to what the employee's problems are and determining how to assess their capabilities so that the AI can better improve them. By the way, the AI does a pretty good job of assessment too. Like in Galileo, it figures out over time what your skills, strengths, and weaknesses are from the behaviors and the interactions and the questions you ask it. So the system actually does some of that. There'll be, so there's massive changes in the whole content business. And, you know, what we're now finding from some of our customers, and we have a, you know, a large pharma client, a large airline, a large, rather, insurance company, almost half to 60 to 70% of the instructional designers may not be needed because this AI stuff is so much more efficient. It still needs humans, but it doesn't need as many humans, which means that there will be roles in performance consulting and in face-to-face instruction and in content sort of architecture as opposed to content design. So there's all that. Then there's the issue of where do we get subject matter expertise from? Now, in most companies, we bought courses from a library provider or an expert, or we went out and developed courses from maybe with a contractor or a consulting firm. There were highly sophisticated courses based on our company's domain. So if you're an oil company, I mean, Shell Oil built entire virtual simulations of drilling. I mean, because it's pretty doggone important that people know how to do that.

Seen some of those, yeah.

Well, you know, that was millions and millions and millions of dollars to do that. You can now gain, you can now capture expertise with AI just by interviewing an expert. Now it's not quite the same, but it's pretty close because what can happen with AI, which we could never do before, is you could interview somebody like me or whatever the domain expert is, ask them a bunch of questions, have the AI do that of 20 experts. By the way, the AI could do that without your help. It could send them all an email and just record their conversation. And the AI will essentially become a digital twin or a digital expert of those people. Now, it's not as sophisticated or nuanced as a human,, but it can gain a lot of information. So if you could think about subject matter expert, sort of collection from all of your repair people, all of your customer service people, all of your managers that you consider to be high performers, all of a sudden you're building a body of knowledge that might have taken you years to do in the past and was very hard to turn into instruction. So that's another huge area of new roles and new, I don't know what to call those jobs, but they're going to be different jobs and they're going to be well suited in L&D as maybe knowledge collectors, knowledge curators, something like that.

I've seen projects like that where deep domain knowledge is captured and then fed into a chatbot for coaching and learning in the flow of work, helping people through it. It was a fairly speculative phase of development. It was a kind of very bleeding-edge project at the time, and things have moved on apace since then. So on the technology side, we're well past the hype phase for AI and learning, I think. There's still plenty of hype about, but where do you see the real durable changes happening in learning? What happened in 2025 and HR tech? And what should learning leaders be investing in or experimenting with as they plan ahead in the next year? We keep the focus kind of limited to the coming year.

Well, I think we're back to the days I remember in the early days of e-learning where you need some experimental investigative learning technologists to start playing with this stuff because it is so different from what we've had before. Digital twins, avatars, online coaches, super tutors. These things are available. You can buy them. You can turn them on. You can use them. But you're going to have to figure out where they fit in your company and where they will best be useful. You know, there will undoubtedly be 100 new technologies in L&D. I'm sure when you go to this conference in Berlin, you're going to see tons of them. In the education industry, but the L&D professionals need to decide which ones make most sense and how to use them in the domains that the company needs the most help. I'll give you an example of digital twin. So we use a digital twin technology. We're sort of more like a professional services company and a research company than anything else. So we have 45 or 50 people, each of whom are experts at various things. Talking to clients all day. So we're generating massive amounts of IP in our domains of expertise and in the understanding of what's going on in the market, whether it be vendors or corporate, you know, HR people. Well, that data sits in people's heads or sits in their emails and goes nowhere. Well, the digital twin's now capturing all of it. So I can go to our digital twin and I can ask it questions. I can ask it operational questions like, what's the status of such and such a contract? Or I can ask it, you know, what is the sum information we've collected so far on the role of frontline management skills in healthcare from the clients we've talked to? Now, you know, that might be a 3-week project if I did it by hand, but it might only take 15 minutes to an hour if I did it with the digital twin. I don't know if that particular use case applies to L'Oréal or Shell? Probably not. But it does to a company that's doing a lot of service work, or it might apply to the customer service department of a big company. Genpact, for example, which is a big outsourcing firm, is using that type of technology to capture knowledge from a client in order to build better services for the client. So right now you need some experiment tech kind of wizards to play with this stuff who really want to figure out how to apply it to the big, best business cases in your company. I would say a year or two now ago, a lot of these tools were kind of primitive ideas and they didn't really work. Now you can buy it or experiment with it, and the vendors are kind of ready to work with you. So that's one thing. And then the second thing is starting to look at these massive corporate course libraries and LXPs and LMSs and scratch your head and say to yourself, how much are we spending on this? What is it costing us to maintain this? And what is the value we're really getting? And make a sober assessment of whether you need it. I think a lot of this stuff can be junked, to be honest. Not overnight, but I think we have one, this large pharma client basically went through a 2 or 3 year effort to re-engineer their L&D function. And they're now at the point, 3 years into it, where they're eliminating, I don't know, 60%, 70%, 80% of the infrastructure they had to replace it with AI. And they're going to save a lot of money and they're going to save a lot of time and their solutions are going to be much faster time to market. So there's the beginning of, let's think about this as a 2, 3-year transformation from old SCORM LMS content stuff to AI native and let's plan it. And because some of it won't come along for the ride, some of it's gonna have to stay where it is. So there's that initiative in L&D too. And the third thing I would say is to be a little wary of is some of the AI vendors are building AI tools to build old style experiences. So, I mean, I don't think courses are going away, but the course paradigm is very tired. It's really kind of old. I mean, I don't really go through— I go to YouTube and I learn things in 15 or 20 minutes, or I come back and I watch them longer. I don't take very many courses anymore. Clicking through chapters and trying to figure out where the content is that I need, that's going to kind of go the going the way of the dinosaur pretty quick here. So I think there's a risk that if you buy something that allows you to more quickly develop old stuff, you're not going in the right direction. So you have to be a little— that's why we want people to read our revolution paper because this is a revolutionary way of rethinking the whole domain.

Going back to digital twins, that is a fairly accessible technology. You mentioned Galileo Summit and our sponsor Synthesia. Yeah, Synthesia do some stuff like that. You can just go in and do it. I've contemplated having my own digital twin made, but my experience of using LLMs is that they're very polite. The guardrails stop them being particularly controversial. In fact, they're quite obsequious, even servile. And I don't think I could get one to ask annoying questions.

You know, John, you're absolutely right. I've actually written about this. I think they're trained to be good citizens and to answer questions even when they don't know the answer.

Yes, of course.

I think you're absolutely right.

That's a good observation. Yeah, but they don't like to go to, I've noticed sometimes, you know, I'll help, I'll ask ChatGPT to help me out with drafting questions and just to kind of lay out a first draft to work from. And if I'm dealing with a subject that has anything in the least controversial, like a guy who wants to talk about immigration, there just won't be a question on it. Okay, so you travel widely and work with organizations across regions. You've recently done some travel, I know. What differences are you seeing between the US, Europe, and Asia, and the Global South perhaps in general you'd like to talk about, in how companies are adopting AI, reshaping skills, and rethinking learning? And what should, should L&D leaders, HR leaders be particularly alert to in the year ahead?

Yeah, I mean, it's very interesting. I've been to Asia recently, I've been to Europe a lot, and I've been to the Middle East very recently. The parts of the world where the economy is newer and growing faster have the opportunity to start from scratch. So if you go to the Middle East, there's some old companies there too, but there's a lot of new companies that are growing very, very fast, and they can take a clean slate and start with an architecture that's AI-centric from the beginning. You can't do that if you're a big old multinational European or American company. You just can't. You have too many legacy systems. So that's one big difference is how old is your infrastructure and how comfortable or ready or familiar are you with changing it? There are cultural differences too. I think in the United States, companies like to disrupt stuff. They like to be highly competitive. They like to do kind of wacky things. There's just a little bit of a sort of an American style of pioneering here. So there's a fair amount of experimentation going here. I find in the European organizations, leaders are a little more methodical. They're more long-range thinking. They're more likely to think about architectures. Perhaps. I know that's a very big generalization. And they're more concerned about the employee issues because they either have labor relations issues or works councils, and they can't just radically change a bunch of jobs without getting everybody to agree. That said, a lot of European companies outperform American companies because they're more patient. I mean, if you look at Airbus versus Boeing, that's a perfect example. Boeing was a typical American company. Chasing its stock price. Airbus has been taking care of its airplanes. So there's probably a lot of examples like that. Then when you go to Asia, it's pretty wild. I mean, Japan is a very old traditional model of work and a traditional model of business. And Japanese companies are slow to change, but very good at long-term change and long-term thinking. If you go to China or Hong Kong, You know, people are very creative, very innovative. You go to Singapore where, you know, where a lot of HR departments are, there's millions of, they're trying, willing to try new ideas all the time. Digital transformation is much more advanced there. The government is much more involved in helping companies. That's another difference. If you go to the Middle East or certain areas of Asia, the government is very involved in helping companies transform. In the US, I would say the government is mostly in the way, to be honest, or maybe not necessarily helping that much. So, you know—

Or maybe picking winners.

Yeah, or picking winners. But you know, the other thing that's funny, the world isn't as big as it seems, at least for me. Everywhere I go, people are more or less the same. They have similar issues. They're all learning at the same rate. There's lots of migration of people around the world working in different geographies who came from different countries. You go to the Middle East and 85% of the population is expats who came from Europe or came from Asia. So it's not like there's only Arabs from the Middle East there. It's lots and lots of heterogeneous people. But I think a lot of this comes down to the company you work for's either tolerance or interest or readiness to adopt these new things. I mean, there's going to be risks with AI, it's not a perfect deterministic technology. Sometimes it doesn't do the right thing and sometimes it doesn't generate the right content. But if you're willing to deal with that and accept that, the upside is just massive. So, you know, how risk tolerant and how risk sort of familiar is your company with this? And we're happy to, we have maturity models for this. So anybody that's listening to this that wants to step through their company's situation, we'll be happy to help you figure out where to start.

All right, sure you will. If you had to pick 2 or 3 signals that leaders should watch closely in 2026, signs of where the workforce and learning ecosystem are really heading, some people do this, I'm not going to get you to make predictions necessarily as such for the year ahead because I know that could be hostage to fortune.

I'm working on that right now.

But what signal? Yeah, Do you have kind of particular signals that you look out for and, oh, there's that one moving?

You know, the big signal or opportunity is to think of AI as an empowering and learning opportunity, not a job destruction, cost reduction, productivity project. Yes, things will become more productive. Yes, there will be jobs that will be eliminated. Yes, you could potentially eliminate certain numbers of people in the company, but that is a very small opportunity compared to the opportunity of delivering information to people faster, training them faster, getting them to be more productive with customers, getting them to know more about the business, keeping them up to date on new policies and procedures, sharing information internally. Those things are 100 times more valuable than laying off 10% of your workforce because you got some automation stuff in there. Because even if you did lay those people off, the odds are they could be doing something else. So, you know, I think the theme to me in the enterprise part of AI, forget about the consumer stuff, is are you thinking about this as growth and enablement and opportunity And are you investing in the skills and technologies and the change to do that? That's why super managers are so important, is managers have to be able to facilitate and educate and enable people to learn and use these new tools and also create stuff. A lot of the creation of AI solutions is coming at the front line because you can create something by prompting it. You don't have to be a software engineer. To program it. So that's, to me, is the big issue in 2026. We could have a recession, in which case, you know, people don't think so much about that. But, you know, that takes patience. You know, during recessions, we've been through some big ones in my career. Some companies ratchet back everything. Some companies look at it as an opportunity to innovate, you know, despite the fact that there's a slowdown. So, I mean, the tools are not perfect, but they're far enough along that you should not be reticent to be involved in this. And you should plot a transformation agenda, particularly in L&D. L&D is behind. I mean, if you think about HR, recruiting people have been using AI for a decade or longer. Employee services call centers have been using AI for a long time. L&D, this is new. We haven't had this level of automation in L&D. Ever. And I think it's the most exciting area of all. And I haven't tried to size the market, but if I think about all of the content stuff that we buy and build and the training and assessment and coaching tools we have, I think this is probably a trillion-dollar market that will be revolutionized by AI. It's probably as big as many of the other areas of of IT and other services. And we're just in the beginnings of it. So, you know, find some creative people in your company that are willing to put a business lens on this and lean into it. And I think you'll be very, very happy in a short period of time.

How do you see the market in terms of LearnTech? You're involved, I think I heard, in the acquisition of Sana by Workday. Is this a sign of consolidation? Do we expect in the new year to see more consolidation? It's one example, but would you expect to see more consolidation like that in the market, or is it going to be kind of a battle between incumbents versus born AI upstarts?

If you look at the big learning providers, Docebo, Cornerstone, Workday, SAP, Oracle, LearnUpon, all the others. Many of them are going to have to make a decision is do they buy an AI technology to transform their technologies and solutions or do they build it? Even to greed. And so I think we're in an expansion phase of the market where there are many, many startups building many, many creative things. I would guess that 80% of them are going to go nowhere, or they're going to find themselves with a product that's kind of a commodity. Unfortunately, because AI is so easy to create stuff, your ability to create a company is about as fast as somebody else's ability to create a company. So there's going to be too many companies, and then there'll be a consolidation. I don't think there's a big consolidation taking place. Yet. I think Sana is a great company and we continue to work very closely with them. We think very highly of them, but there will be other companies that will compete with them. There already are, that will either be born or grown from the technology that's out there. And so we're not at a, we're not at a consolidation phase. We're really at an expansion phase. Like I said, if there's a recession, Or the AI winter comes back and there's an AI crash, which everybody keeps talking about. Well, maybe that'll happen, but we're certainly not there at the moment.

And finally, the traditional Learning Hack Closer. What's your go-to karaoke song? We're of an age, so it's gonna be something I've heard of.

I am not a singer. I'm sorry. I do not do karaoke. I've tried it. It's so embarrassing for me. It's just not my thing.

Sorry, John. I won't press on that. Okay. Thank you very much. It's been a really interesting discussion for me, and I think I did get the chance to ask you some quite rude and impertinent questions, which ChatGPT wouldn't have done. And thank you for suffering that with good grace.

Thanks a lot, Geoff. My pleasure, John.

Thank you. That's all on the learning podcast for this time. Many thanks to our guests and to our sponsors. The Learning Hack is completely independent and transparently funded by sponsorship. It is produced by Jon Helmer and hosted by him and also me, Mariana, a Synthesia AI avatar. If you want to help others find us, please like, follow, rate, review, and subscribe wherever you get your podcasts or on YouTube.

Will be back in your feed in 2 weeks' time.

Meanwhile, stay curious, learning people.

Now I finally get it.

The Learning Hack Podcast is supported by Learning News, the learning sector's newswire. Rob and his team are good friends of the podcast, and we really value the help and advice we've had from them, and they do a great job. For the very latest news from around the learning sector, for interviews with learning leaders, the latest from learning sector vendors, and features on workplace learning, go to learningnews.com.