[00:00:00] You know, the organizational psychologist Adam Grant recently dropped a quote that I just, I really can't get out of my head lately. Oh yeah, which one? He said, great careers used to hinge on ability and now they depend on agility. Ah, yeah. That's a powerful way to frame it. Right. And he wasn't talking about like, just keeping up with the latest software update on your phone. He was making a much deeper point about how in a world where information is basically a cheap commodity,
[00:00:27] the specific expertise you hold today might just be completely irrelevant tomorrow. Exactly. The real differentiator isn't what you have memorized anymore. It's how incredibly fast you can learn, how you adapt, you know, and how you exercise judgment in situations you've never actually encountered before. It totally forces a redefinition of what professional value even means. Yeah. We're stepping away from this old school idea of the static expert, the person who has all the answers,
[00:00:55] and moving toward the concept of the continuous learner. Yeah. And that shift is exactly what we're unpacking in today's deep dive. We're looking at a fascinating article by Dr. Jonathan H. Westover. Right. And he's exploring how the combination of two specific things, what he calls micro experiments and artificial intelligence, is totally rewiring how we learn to become leaders. We've got a great stack of sources for this today, too.
[00:01:20] We're pulling from experiential learning theory, cutting edge AI organizational research, and some real world case studies from places like Microsoft and the Boston Consulting Group. So our mission today is to figure out how this tech and psychology blend actually works for you. But to frame the core problem we're solving, I want to borrow an analogy from the source material. Think about swimming. Yeah. You could sit in a quiet classroom for a week and read books about the biomechanics of the freestyle stroke, right?
[00:01:48] You can study fluid dynamics, breathing techniques. You could even ace a multiple choice test on it. Sure. But doing all of that makes you a better swimmer in the exact same way that reading about a leadership concept like, say, psychological safety, makes you a better leader. Which is to say it doesn't at all. Exactly. If you jump into the deep end with only theoretical knowledge, you're going to thrash around, swallow water, and probably sink.
[00:02:12] That is the perfect visualization of what researchers Pfeffer and Sutton call the knowing-doing gap. There's just this massive distance between having an intellectual insight in a classroom and actually enacting that insight in the messy, chaotic reality of the workplace. Well, wait. If that's true, and it intuitively feels true, why are companies still pouring, like, billions of dollars into traditional training?
[00:02:38] If we're spending millions on these fancy week-long leadership retreats, are you saying it's mostly a waste of time? Well, I mean, the research from Beer and his colleagues suggests it often is. The problem is that the inspiration from a great workshop rarely survives the return to the office on Monday morning. Why doesn't it survive, though? People leave those retreats so pumped up. It comes down to cognitive load and how the human brain processes stress.
[00:03:02] You see, in a quiet retreat center, surrounded by your peers and drinking good coffee, it's very easy to declare, I'm going to be a more inclusive listener. But then Monday morning hits. You have back-to-back meetings, a crisis with your biggest client, and an overflowing inbox. Under that kind of high cognitive pressure, your brain becomes ruthlessly efficient. Meaning it falls back on old habits.
[00:03:25] Exactly. It shuts down the prefrontal cortex, the area responsible for new, deliberate learning, and defaults to the basal ganglia, which houses your most deeply grooved automatic habits. Wow. So if your habit for the last 10 years has been to just interrupt people and dictate solutions to save time, that's exactly what your brain forces you to do under stress, regardless of the certificate you just got at the retreat? You'll do it on autopilot. It's the difference between a conscious intention and ingrained muscle memory.
[00:03:53] And this failure to translate knowing into doing has a massive financial cost. How so? Well, when a leader goes to a seminar on coaching, but then defaults to micromanaging the second a project gets delayed, it severely erodes team trust. We see in Gallup's data that managers account for a huge share of the variance in employee engagement. Which directly drives profitability. Right. So when leaders default to bad habits under pressure, the whole organization bleeds potential.
[00:04:22] Which really makes sense of this 70-20-10 framework that the Thorses talk about. The idea is that 70% of adult learning actually happens on the job through real experience, 20% happens through feedback, and only 10% happens in formal instruction. And yet corporations spend almost all their budget obsessing over that 10%. Yeah, it's so backward. And it gets even more glaring when you look at the time constraints on modern workers.
[00:04:49] Josh Burson's research shows that employees only have an average of about 24 minutes per week available for formal learning. 24 minutes a week. That's what? Less than 5 minutes a day? Barely enough time to watch a single training video. And the crazy thing is, while most executives say learning is a top priority, only about 12% of organizations actually deliver learning effectively in the flow of work. 12% is a staggering failure rate.
[00:05:12] It really proves that if traditional workshops can't survive contact with reality, the solution has to be embedded directly into daily work. We can't separate the learning from the leading. So, because those isolated workshops fail, we need a behavioral solution. And that brings us to the core concept here. The micro-experiment. Right. Researchers like Christopoulos and Watkins define micro-experiments as small, specific, observable acts of trying something different in a real work situation. Okay.
[00:05:42] And here's the crucial psychological distinction. They are driven by curiosity in data collection, not by the demand for a flawless outcome. I want to push back on this terminology just a bit because, at first glance, a micro-experiment just sounds like a fancy corporate buzzword for a goal. What's the actual difference? Let's say you're a manager and you realize you're a bottleneck for your team. Okay. A common problem. Yeah. So, a goal would be, I'm going to delegate more effectively this quarter. How is a micro-experiment different?
[00:06:10] A goal like that is a noble sentiment, but it gives you absolutely no scaffolding for how to act in the actual moment. It requires immense willpower exactly when you're most stressed. A micro-experiment takes that vague desire and turns it into a specific bounded action. So, a micro-experiment would be something like, in tomorrow morning's project meeting, when we get to the action items, I'm going to ask the team to propose who owns each item and observe my own comfort level while I wait for them to speak. Yes.
[00:06:41] Notice the shift in mindset there. You're no longer demanding perfect delegation from yourself. You're running a test. This taps right into Kolb's learning cycle. You have the experience, you reflect on it, you conceptualize a new approach, and you experiment again. It also leans heavily into Carol Dweck's research on growth mindset, doesn't it? Oh, absolutely. When you frame an action as an experiment, you're adopting a learning orientation rather than a performance orientation. You aren't trying to prove how good of a boss you are.
[00:07:10] You're just tweaking a variable to see what happens. Okay, let's unpack this, though, because I feel like this is where people get hung up. Sure. Isn't it incredibly risky to run experiments on your actual team in a high-stakes corporate environment? What if that delegation experiment completely bombs and the meeting devolves into 10 minutes of awkward, dead silence? That's the beauty of making the experiment bounded and specific. The primary goal is data collection about your own leadership context. So it's okay if it bombs. Exactly.
[00:07:39] If you run that test and the room goes dead silent, you didn't fail. You gathered incredibly valuable data. You just learned that your team is currently so conditioned to wait for your orders that they don't feel comfortable volunteering. Ah, I see. Yeah, so a failed interaction is actually a successful experiment if it gives you the insight you need to recalibrate your approach for the next meeting. It totally takes the pressure off. But there's a serious friction point here.
[00:08:07] If these micro-experiments are so effective at bridging the knowing-doing gap, why isn't everyone already doing them? Well, translating a broad, fuzzy insight like, I need to empower my team, into a highly specific behavioral test takes a massive amount of cognitive effort. Right. Usually you'd need a highly trained executive coach to sit down with you, listen to you vent about your week, and help you design the experiment. And that level of personalized coaching is incredibly expensive and time-consuming. It doesn't scale.
[00:08:37] Which creates the perfect entry point for artificial intelligence. Let's really dig into this tech side. Because AI's role here is fundamentally about translation, right? It's not just a chatbot giving you a top 10 listicle on leadership. Exactly. The Center for Creative Leadership look at how AI-powered tools help leaders articulate their challenges. And they found the AI is remarkably good at reframing problems. What does that interaction actually look like in practice?
[00:09:05] So imagine you log into one of these platforms, like the Permio system, developed by the Leading Through Institute. You tell the AI, my team isn't taking initiative on this new project and I'm frustrated. And the old way of doing things would be just searching a corporate database for articles on team initiative. Right, which you don't have time to read anyway. But the AI doesn't do that. Instead, it offers a coaching prompt. It might ask, what about this challenge will require you to grow or adapt? Oh, wow. That is a powerful pivot. It really is.
[00:09:35] It forces you to stop complaining about your team's shortcomings and look in the mirror. It shifts the framing from an external annoyance to a personal developmental opportunity. And once you make that shift, the AI guides you through designing the microexperiment. Yes. It helps you figure out exactly what to test in tomorrow's meeting. It might even roleplay a difficult conversation with you, playing the part of a defensive employee so you can rehearse your tone. That's incredible.
[00:10:03] And then crucially, it pings you after the meeting to prompt your reflection. How did it go? What did you feel when they pushed back? I read that the Boston Consulting Group's capability building arm, BCGU, is leaning hard into this. They're trying to move learning out of the classroom almost entirely and make applied learning the core of how they build skills. Yeah, they're using AI to meet managers where they actually are, right in the trenches of their daily schedules. Here's where it gets really interesting for me.
[00:10:32] Is the AI basically acting like the bumpers in a bowling alley? Oh, I like that. How do you mean? Well, it's not throwing the ball for you, right? It can't run the meeting for you. It can't read the body language in the room. And it certainly doesn't feel the anxiety in your stomach. But it's keeping your experiment out of the gutter. That is a brilliant way to conceptualize the division of labor here. The AI provides the scalable, structured scaffolding. It gives the nudges. It forces the reflection. It tracks your patterns over time.
[00:11:01] It acts as the bumpers. Well, the human does the actual rolling. Right. The human being retains all the agency and provides all the emotional and relational depth. Only a human can exercise empathetic judgment and build real trust. The AI just ensures you actually take the time to reflect on that human interaction afterward. Okay, so AI gives me the bumpers to roll the ball. But if I think the owner of the bowling alley is going to kick me out the second I roll a gutter ball, I'm never walking into the building in the first place, you know? That's the catch.
[00:11:31] The technology and the experiments don't exist in a vacuum. If a manager is terrified of making a mistake, no AI prompt is going to get them to try a new behavior. Which is why the surrounding environment is just as critical as the technology itself. We have to talk about Amy Edmondson's concept of psychological safety. She defines that as the shared belief that a team is safe for interpersonal risk-taking, right? Exactly. If an organization lacks that safety, this entire AI-driven experimental system collapses instantly.
[00:12:01] Think about your own workplace for a second. If you, the listener, tried a completely new experimental way of running a meeting and it totally flopped, would your boss applaud the learning or penalize the mistake? That is the million-dollar question. Because that determines whether this whole system can actually work for you. Right. If you think a messy experiment is going to be weaponized against you in your performance review, you're not going to do it. Of course not.
[00:12:26] Your brain senses a threat and the only logical incentive is to revert to your safest, most familiar patterns. You'll never try anything new. If we connect this to the bigger picture, organizations have to explicitly reward learning agility, not just flawless execution. The classic example in the research here is Microsoft under Satya Nadella. Yes. When Nadella took over as CEO, he recognized that Microsoft had a toxic know-it-all culture.
[00:12:53] Everyone was terrified of looking stupid, so everyone had to prove they were the smartest person in the room with all the perfect answers. And he explicitly championed a shift to a learn-it-all culture, leaning on Dweck's growth mindset. Exactly. That top-down mandate provided the psychological air cover necessary for people to experiment, to admit when the strategy failed, and to adapt quickly. And the research highlights that this isn't just about feeling individually safe. It's about the social ecosystem. Yeah.
[00:13:21] Vince's research on group interaction is really relevant here. Right. He talks about the importance of cohort-based accountability. When you build programs where leaders are forced to share their successes, and especially their totally awkward failures, with a group of their peers, it enriches the sense-making. Sharing those awkward failures is vital. The AI can help you structure your personal reflections, but making sense of the nuance, like why a specific phrasing fell off, that requires human community.
[00:13:48] It turns isolated data points into collective organizational wisdom. So when you combine all these elements, bridging the gap with micro-experiments, scaling it with AI scaffolding, and anchoring it in psychological safety, you don't just change how leaders learn. You fundamentally change who gets to learn. Historically, if you wanted intense, experiential, personalized coaching, you had to be a senior executive. The unit economics of one-on-one executive coaching are just staggering. Right.
[00:14:18] You're paying a coach hundreds of dollars an hour. Yeah, you can't provide that level of investment to a 26-year-old frontline shift manager. It was an elite privilege. But AI changes the unit economics of reflection. It democratizes the entire process. It brings that continuous cycle of bounded experimentation all the way down to mid-level and frontline managers. And those are the people who manage the vast majority of the global workforce. It's shifting the entire foundational layer of leadership. And the ultimate goal here isn't just downloading a new tactical skill into your brain.
[00:14:48] The researchers lean on John Dewey's philosophy of building a dynamic learning identity. It's this idea of the continuity of experience. When you run 20 micro-experiments over six months, you haven't just learned 20 isolated party tricks. You've fundamentally shifted how you view yourself. You no longer see yourself as a static expert who knows how to lead. You see yourself as a dynamic practitioner who is constantly learning. The leading-through institute maps this as a clear progression.
[00:15:16] It starts with no getting the initial insight. Then it moves to feel and do cat, actually running the bounded experiment and processing the friction in the real world. And ultimately it leads to become. Exactly. Where adaptive reflective leadership isn't just something you do, it's who you are. So what does this all mean? Does democratizing development down to the front line mean the end of the exclusive high-paid executive coach? Are we replacing humans with AI? Not at all. It's an expansion, not a replacement.
[00:15:46] The human-AI partnership ensures that the connective tissue of learning is maintained over time for thousands of leaders simultaneously. It's like a tiered system. Right. A first-time manager gets immediate scaffolding from the AI. Meanwhile, a senior executive might use that exact same AI to track their behavioral experiments during the week so they can bring richer data into their session with a human coach. The AI handles the massive scale, and the humans handle the relational depth.
[00:16:15] It is such a wild paradigm shift. We started by looking at that frustrating gap between knowing what to do and actually doing it. Why jumping in the pool is harder than reading the book. Exactly. We explored how bounded micro-experiments can bridge that gap by giving us real-world data. We saw how AI acts as the bumpers in the bowling alley, translating vague goals into structured tests. We anchored it all in psychological safety and saw how this democratizes development across the entire corporate ladder.
[00:16:44] This raises an important question, though, one that the source material hints at, but the industry hasn't fully resolved yet. Oh, what's that? If we're building these continuous learning systems where an AI tracks our micro-experiments, our reflections, and our behavioral patterns over months and years, what happens when the AI understands a leader's blind spots and growth trajectory better than they understand it themselves? Oh, wow.
[00:17:10] If the AI has the receipts on every behavioral experiment you've run for five years, and exactly how you adapted or failed to adapt, how will that change the very nature of performance reviews? That is an entirely different deep dive right there, a transparent AI track performance review that knows your habits better than you do. It's terrifying and brilliant all at once. But for now, we want to challenge you, the listener, don't let this just be another piece of static knowledge you consume today. Absolutely. Think about one leadership insight you've been struggling to implement at work,
[00:17:40] and design one tiny bounded micro-experiment for tomorrow morning. Get out of the classroom, get in the water, and just see what happens when you swim. Thanks for joining us on the deep dive. Thank you.
Sponsors
