E110 - When AI Enters The Systems Room - Podcast with Joan Lurie

Welcome to The Inner Game of Change.  where we explore the thinking that shapes how change really happens. 

Four years ago, Joan Lurie joined me on this podcast to explore systems, roles, mental maps and how organisations adapt in times of uncertainty.

Today, we return to that conversation through a different lens: artificial intelligence.

But this is not a conversation about tools, prompts or technology.

It is a conversation about what AI is revealing. The assumptions it is challenging. The roles it is reshaping. And what happens when AI enters the systems room.

Joan is the founder of Orgonomics and one of Australia's leading systems thinkers. She has a unique way of helping leaders see organisations not simply as structures or collections of people, but as living systems of relationships, roles and patterns. I have learned a great deal from Joan's work over the years, and every conversation with her challenges me to see organisations differently.

Together we explore leadership, adaptation, complexity and what organisations may need to rethink as AI becomes part of the fabric of work.

I hope you enjoy this thoughtful and timely conversation with Joan Lurie.

About

I am the CEO of Orgonomics, a company I founded in 2008 to assist leaders and organisations to adapt, thrive and function at their growing edge. My work integrates strategy, systems thinking, complexity, narrative, ecology and developmental theory. 
 
Over 30 years I’ve worked with boards, executives and leadership teams as a coach and consultant helping them achieve turnaround results - together emerging new cultures, operating models and different organisational systems, whilst simultaneously building their adaptive capacity. 

Through this practice and experimentation, I have developed Orgonomics™ - a proprietary methodology which provides a ‘map’ for leaders to go beyond reductionism to navigate the complexity we are in. 

Referred to as ground-breaking, Orgonomics enables leaders to fundamentally shift their ways of seeing and knowing to be more systemic; reframe their assumptions and mental maps and repattern their organisational systems for new ways of relating and operating, to achieve higher order functioning and performance.

As a developmental psychologist and systemic change consultant my core purpose is to ensure that leaders and organisations are able to continuously develop, learn and grow to be fit for and thrive in our complex world. Developing a systems lens, relational intelligence and comfort with not ‘knowing’ used to be a ‘nice to have', but it has now become our individual and organisational imperative; we have to accelerate it for the good of our whole ecology.

Contact

Joan’s profile

linkedin.com/in/joan-lurie-73bb0215

Website

orgonomics.com (Company)

Transcript

AI As A Catalyst For Seeing

SPEAKER_03 0:00

I think we've always seen traditionally, and it's still the dominant logic, organizations as sort of technical structures, organization charts, processes, reporting lines, sets of capability, or we've seen them as collections of people, and that the way that we improve them is through improving people, growing people, developing people, and improving interpersonal relationships. In fact, if we look at the dictionary definition of what is an organization, the majority of dictionaries, there's a kind of standard description, which is an organization is a collection of people with a common purpose. With the emphasis being on collection, I think that these two lenses, the technical and the sort of interpersonal psychological, have dominated the way we've made sense of organizations. But of course, there's a third way.

Ali 1:15

Four years ago, Joan Lurie joined me on this podcast to explore systems, roles, mental maps, and how organizations adapt in times of uncertainty. Today, we return to that conversation through a different lens, artificial intelligence. But this is not a conversation about tools, prompts, or technology. It is a conversation about what AI is revealing, the assumptions it is challenging, the roles it is reshaping, and what happens when AI enters the systems room. Joan is the founder of Orgonomics and one of Australia's leading systems thinkers. She has a unique way of helping leaders see organizations not simply as structures or collections of people, but as living systems of relationships, roles and patterns. I have learned a great deal from Joan's work over the years, and every conversation with her challenges me to see organizations differently. Together we explore leadership, adaptation, complexity, and what organizations may need to rethink as AI becomes part of the fabric of work. I am grateful to have John having this conversation with me today. So, John, so good to see you, and thank you so much again for joining me in the Inner Game of Change podcast.

SPEAKER_03 2:44

So great to be back and reconnect. Thank you, Ali.

Ali 2:47

Thank you very much. In September 2021, though we recorded an episode and we spent most of the time talking about systems roles, mental maps, and at the time we were navigating the pandemic. And four four years later, yeah, five now almost, we find ourselves in another wave of change, which is artificial intelligence. And I I looked at the notes from the last podcast and um I found a lot of similarities. And I today I think we're gonna go a little bit deeper into what that seismic change. We are in the early stages of it, but I think this can be really a useful conversation. Looking at this, it made me wonder whether AI is not simply introducing change, but maybe helping us see aspects of the organizations that were always there. There was an image the other day that I was thinking about. I was walking by the Yara. One of the topics that I talk about is that AI has got this uncanny capability now to expose vulnerabilities and things that we haven't been able to see before. It's like a river when it recedes, only when it recedes, you can see you know the rubbish, the rocks, and all of these things, but before we couldn't really see them. How are you seeing this change at a higher level

Three Lenses On Organisations

Ali 4:20

now?

SPEAKER_03 4:21

Hmm. Yeah, I think it's such a fascinating question, Ali, and it really makes me think. I I think one of the things that I have spoken a lot about, particularly over the last few years, is that the way that we see and make sense of organizations has to be reframed. And the I think what AI is bringing to the fore is this new conception or different conception, not necessarily new, that we need to start to hold about what is an organization. I think that a lot of the meaning making we've had about organizations historically and the dominant logic about them has been obfuscated or not seen because of the lenses that we've brought, because of the filters of how we've made sense of them. I think what AI, from my perspective, is really bringing to the fore is that we need to challenge those lenses, remove those filters so we can really see organizations through this new and different lens. And so what are those lenses, I guess, is that the I think we've always seen traditionally, and it's still the dominant logic, organizations as sort of technical structures, organization charts, processes, reporting lines, sets of capability, or we've seen them as collections of people, and that the way that we improve them is through improving people, growing people, developing people, and improving interpersonal relationships. In fact, if we look at the dictionary definition of what is an organization, the majority of dictionaries, there's a kind of standard description, which is an organization is a collection of people with a common purpose. With the emphasis being on collection. And I think that these two lenses, the technical and the sort of interpersonal, psychological, have dominated the way we've made sense of organizations. But of course, there's a third way to see them, which is that they are complex networks, ecologies of roles, role relations. And in fact, that where behavior emerges from is from that network, from the system of relations, and that it doesn't reside in individuals or in individual parts or even a collective of those. You know, we hear a lot about the move from R to We, but what that kind of doesn't talk about is the interrelations between and the patterning. So I think what AI is going to force us into is to hold this third lens about organizations. And yes, we can see them as technical structures. Yes, we can see them as collections of people, but really we need to see them as complex ecologies or systems. Why is that coming to the fore? Because so much of what's going to be we're having to navigate is agents with people. So we can't simply say anymore that organizations are collections of people with a common purpose, because that is almost emerging as not true.

Ali 7:49

What is the impact of all of this change on the way leaders look at their organizations?

SPEAKER_03 7:58

Well, I think if you look at an organization, if you hold an assumption and you look at your organization as a collection of processes and structures or collections of people, the kind of linked follow-on assumptions are that the ways that we fix them, change them, optimize them are through using those mechanisms, technical mechanisms or interpersonal psychological mechanisms. So, you know, we can, if I go and talk to leaders and I say, how do you understand the challenges you have? They will usually 99% of the time describe me the challenges in those two kind of buckets based on those assumptions. And their solutions look like. We'll fix it through better communication, we'll fix it by building trust, we'll fix it by restructures. But actually, what the implication now is for leaders is that those two sets or of levers are not going to be sufficient. Because you can't simply now say, we're going to improve it by improving interpersonal relationships or building trust. Actually, the implication for leaders is that they have to be system designers, network designers, and also coherence builders, where they have to start to look at the integrated system of agents and people and how does this complex network of roles actually work? Because what AI is doing is it is challenging every role in every organization. There's not a single role, not one of us, who are not implicated by AI in terms of how it's challenging the role that we play or challenging the way we take up a role. So the boundaries of role, how what they are, how they interrelate, that's where the new opportunity lies.

First Order Vs Second Order Change

Ali 9:50

Is it okay for me to start thinking about the nudge that AI is actually causing? Is it a nudge to the workflows or is it a nudge to the roles? Because an organization is a a system before a human joins that system. Absolutely. So it it and and when it impacts a workflow, the the downstream impact will be an impact on individuals' identity, their knowledge, their capabilities, what originally they were hired for, all of these things. Am I looking at this sort of a it's almost like a first and a second order impact?

SPEAKER_03 10:35

Yes, I love that distinction. I think that leaders and organizations have to look through the first order and second order lens. And if we use kind of my frame, the technical versus the kind of systemic, I think the you could kind of technically look at your workflow and look at how what's the process, what's the flow of work, how does work get done, and where do you put agents into that workflow? The second order is how everybody's roles are implicated and what that means for the individuals who have held roles, who have mental maps or constructs of roles, and how they relate into the organizational network and into how they relate with other humans or with other agents. That's the second order. It really means that it's not enough to map the workflow. You have to look at the second order implications for how the network of roles functions and how individuals function in that context now.

Ali 11:42

How challenging is that for a leader or a middle manager to start thinking of all of these things? Because that is a capability. I mean, system design is a full-on capability. From my experience, when I talk to clients and management people, it's almost like this is too far in the future, therefore, it is abstract, and therefore I don't want to focus on it now. I just want to focus on giving people licenses, which is the first order impact, and see what happens. We're gonna throw more training and see what happens. For me, it's almost like it's my responsibility to understand the downstream impact of this. And therefore, even if I don't have the capability and the way to think about a system, maybe I need to start thinking about who can help me do that.

SPEAKER_03 12:34

Yes. I think that this what AI is doing is it's completely reframing the role of leader. And I think that leaders historically have either been technical experts, good at what they do, enrolled and given jobs because they know what they're doing technically, or they, and we hear this all the time, right? Or they move up the hierarchy, they start to manage teams, and then they have to build a good capability and people leadership. You know, so you hear all the time, are they good people leaders? Can they manage people? And so that sense of leadership being either expertise, knowing solutioning, or managing people, usually vertically, you know, teams below them, now actually the role of leader is to become system leaders, to be able to be coherence builders horizontally across a network of role relations, to know how to hold the complexity of continuous adaption, because as AI comes more and more online and people are using it more and agents are coming in, this network of role relations is going to be have to be held, designed, and held more coherently. So the role of leader is that now. And do leaders have that capability? No, because where we've placed our emphasis is on technical expertise and people leadership. But we're no longer managing, leaders are no longer managing a collection of people, even teams of people, they're managing networks of agents and people down the line. And the sooner they start to build a capability in doing that and have a shared language, a shared set of frameworks, a shared set of processes, even for doing that, leaders who get ahead of that, organizations that get ahead of that, I think are going to be able to thrive and survive down the line. But I think if we continue to hold an assumption that AI is a tool that you can give people and that you'll get the productivity and efficiency benefits by some of the parts, you know, that already is proving to be faulty because already people are not seeing the productivity benefits, and they're seeing actually escalation in cost in the investment. So if you really are going to say we need to both get the benefits from AI, some efficiencies and gains, which is where leaders are wanting to go. Actually, they need to think about it more as how do we build coherence networks, coherent networks where there's flow and where capacity is unleashed, and how do we start to build that? I think there are frameworks, there is language and capability that can be built and shared. I just don't think that's where people have started leaning in as the primary place.

Efficiency Or Coherence With AI

Ali 15:38

You mentioned a couple of things, and I just want to walk you through some experimentation that I was I'm always fascinated by anticipating what the second and even third order impact of change. And I was thinking if the future is gonna be a manager managing a number of humans plus agents, their job is not gonna be easier. In fact, I actually simulated a manager just managing 20 agents. And I went through the process and I discovered that I I am I lose more sleep managing a number of agents than a number of humans. Yes, they do give me the efficiency and the productivity and the speed, but as a result of this experiment, I started thinking maybe AI was not really created. Maybe speed and productivity was not really the ultimate objective of AI. Maybe it is it helps us the way we think about work, the way we think about life, the way we make better decisions. Uh these are far better outcomes of utilizing AI. Am I being pessimistic about this? You know, critique my thinking in here.

SPEAKER_03 16:57

Yeah. I don't think you're being pessimistic. I think that if I I think what you're raising is interesting. I think if people purely see AI as an efficiency mechanism, then we're gonna lose the benefits of it. And you know, it's it's a typical kind of response, which is how do we optimize? How do we get efficiency rather than how do we, you know, create more coherence, how do we grow and develop and you know, create more capacity and creativity with AI. That that is where I think we'll get more of the benefits, but I think where people start is in the polarity. I think we've got to hold the and with AI. I think we can say, yes, there will be some efficiency benefits and speed, but actually they're also gonna be in that the adding more complexity, which is actually potentially gonna slow us down. So if we don't hold an and with this and say it's not purely an optimization mechanism or an efficiency mechanism, it's actually teaching us how to hold more complexity, how to become better at interrelational management. Because how do we manage complex systems? How do we manage relational interfaces? How do we build governance infrastructure for complexity, for adaptation? That's what AI has come to teach us, I think, right? Because putting agents into a system is driving up complexity even more so. And, you know, up till now, I think people talk about, oh yes, complexity, you know, how it could benefit us if we get better at that. I don't think it's like how it could benefit us. It is an absolute imperative. The more you put in the efficiency mechanisms and use agents, the more you actually need the ability to manage the complex network that the agents are creating with you, right? And learning how to have hybrid ecologies and manage those hybrid ecologies and coexist, that's a new complex adaptive challenge for us as a species.

Ali 19:18

Where do you see this happening? Uh what can you see from your interactions now with the clients when it comes to systems? The way they I'm looking at this is that even if you want to optimize an existing system, you are actually optimizing a system that is based on past assumptions. It always reminds me of the electricity story when electricity was introduced, the majority of the factories at the time were based on were designed based on steam engines. And so when they introduced the pro electricity to an existing system, the system was not really optimized. There was not a lot of productivity. The productivity only was was gained maybe a decade later when they started redesigning the factory floor, when they started redesign when they started building a new factories, optimized for electricity. So the fact that we have an existing system now, an organization or a team, and we're bringing in our AI, it'll be it'll be really hard work to try to get the best out of AI on a system that has existed for a previous assumption.

SPEAKER_03 20:40

Yeah, I couldn't agree more. I think the I think that's sort of what I was saying in the opening, that all our assumptions about organizations, how they function, and how we need to work with them, change them, has to be disrupted in order to integrate AI into this organization. I think if we hold those old assumptions about what is effective work and how we're gonna be successful, and we try and drop AI into those existing organizational assumptions, we're gonna get in we'll be in trouble. And I think we're already seeing that. I think people are investing a lot in trying to, you know, in a first-order way, as you say, drop AI as a tool into an already existing kind of paradigm. Whereas actually the paradigm has to shift in order to integrate AI and the agents effectively, I think. And so what are those new assumptions we need to hold? What are those new practices, even about work design, right? I was with a client the other day and kind of posed the question to them if you were going to use AI and improve, what measure? Mechanisms or levers would you need to look at in relation to the AI that you're introducing? And the they clients can spit that stuff out. KPIs, the behaviors, you know, all the traditional things that we have thought of as key levers. And I said to them, well, you could think about it like that, or you can think about it through a systems lens of where are the boundaries being drawn? What boundaries do you need to reset? What are the rules of engagement for how the AI human interface works? What are the rules of engagement for how roles interface now? What mental maps or constructs are people going to have to hold about what effective work looks like and being able to see their role in the system? What about things like capacity for relational forms between agent to agent, human to agent, and human to human in this new context? How are you going to design for those things? So you can kind of hold those two as separate constructs side by side, and you can say, actually, you can't rely on KPIs, behaviors, all of those things alone anymore. But those are the dominant levers that organizations have pulled historically, right? But they need to hold a new set of assumptions about what a system design looks like. What levers do we have available to us? And they're much more about interrelational infrastructure and infrastructure for adaptation, which we don't have built into our assumptions or

Incubation Labs Over Linear Rollouts

SPEAKER_03 23:42

into our practices.

Ali 23:43

You remind me of um a story you probably heard about it. The British government in India. The snakes. The snakes, the cobra, yeah. So the system is really, I mean, I'll probably gonna talk about it, is that they they do a lot of cobras and snakes and all of that, and then so they started building an incentive program for people to bring cobras and they pay them money, and then all of a sudden they realize that some naughty people started breeding more cobras. And so the system is always de optimized to produce how the system is designed for. And so the current system is designed to you know is it's designed based on incentives, workflows, and all of these things. The fact that we need to rethink all of this system, I cannot but think that it needs to be sorry, a sequence of events that an organization or a leader would need to go through. And so sometimes I try to simplify for my clients about what that looks like. Yes, start, for example, generative AI, start with a number of licenses, train your people, then observe the system, then experiment with some new ideas, introduce a few agents, then observe the system, then augment a little bit here and there, then observe. Only then you'll start gaining that design or redesign ambition of oh, maybe I don't need to do all of this in this way, but that only happens when you start gaining insight from the floor. So I try to simplify the language. In your opinion, am I simplifying it too much that I am devaluing the effort, or how how do you think of my thinking?

SPEAKER_03 25:44

Yeah. No, I think that I I would agree. I mean, I you know, it's almost what people have to set up is like an incubation lab around AI and make some very, you know, I think most organizations we can say have given people access to co-pilot or whatever those sort of to use them individually. I think what leaders need to do now, to your point, is that they need to think about strategic places where they're going to try and introduce AI and run those as incubators or labs or places to learn rather than this kind of scatter gun piece that's that's happening. I think the the challenge with that is that many leaders are trying to do that through like traditional transformation program means. And I think what you're highlighting is that one of the assumptions leaders have to give up is how to do transformation. And so how you introduce AI and experiment with it is to create these incubation spaces where you test and learn, test and learn. And they're much messier than traditional transformation programs where you put a team together at the center, they design the solution, and then you roll it out. You know, those ways of doing change and transformation are no longer gonna cut it in this kind of environment because you can't start with the solution up front, right? You've got to design your way forward, test and learn, test and learn and experiment. Exactly what you're saying. So I just I think people don't yet have a mechanism for that because we I see organizations still very much steeped in linear transformation rollouts. And so this is a new capability to build is how do you incubate, test and learn, and how do you pick a few of those to do, and how do you use those as learnings where they can be extrapolated? So I think one of the things I'm seeing is you know, I had a client where they designed the solution using AI, and then they did the workflow mapping, and then they looked at the role implications. And and then they said, Oh, really, we should have done this up front.

Ali 28:19

Yeah.

SPEAKER_03 28:20

So people don't have tools for doing like system observation, system diagnostic, anticipating how the role ecology is going to potentially shift, not even foreseeing that. So I think if we can help leaders build this way of seeing up front and have them almost flip. Like, don't design the technical solution or think about it and then think about how you're gonna roll it out, but rather look at the current system of relations, the roles, the dynamics, and how that's gonna be implicated should you put tech in. Then you can do all of those simultaneously. But it is messier, it's messier work, right?

Ali 29:07

It is messier work, and which means the you know the traditional ways of managing change will all be out of the window. It's an ongoing process. The problem with this is that, or the challenge, is that m many teams are driven by incentives and targets. And here's me and John talking about you need to have this patience and you know, create a space for experimentation that is a longer game gain than you know the next month and all of these things. So I do have a level of professional empathy for leaders and managers to start thinking about well, how do I manage both? And and then you see this also group of clients that are sitting doing nothing now because for them it's like this is too hard. Oh, the licenses are too expensive, and all of these things, too expensive in comparison to what? I think I do not envy leaders and where they are now, but I do strongly believe that there's a lot of information out there in the market, and also the market is not making it easier for organizations, there's a lot of noise in the market. And the other the other variable that is actually causing a lot of trouble as well is that technology itself is accelerating on a daily basis. So I used to you know onboard people into you know Chat GPT and and and Copilot for the last two years. Just in the last six months, the technology just completely redesigned its own ways of working. And which takes me back to your idea of I think you mentioned the term continuous adaptation. But is that underpinned by obviously are we moving from to an organization, a learning organization rather than a doing organization?

SPEAKER_03 31:15

Well,

Building Reflective Adaptive Systems

SPEAKER_03 31:16

I think that the term I think continuous adaptation is what we need, right? I think those sort of days where we could have designed the operating model, put it in place, you know, even if we did the systemic work to get the rewiring and re-pattering interconnection happening, you could live with that for a period of time. I think now we can't rely on that anymore. So the new operating model is how in certain places do we continuously, you know, adapt. I think the challenge with using terms like learning versus learning organization is that it was used by Peter Senge in the fifth discipline. And I think that you could say that organizations could have been learning organizations in other ways in a more technical description. So, you know, there are many organizations that would run plan, do review cycles, they would look at improvements, but much more through a technical lens, but they would have seen themselves as learning organizations, organizations that were continuously improving. So I would say a learning organization is more in that description for myself in terms of how it's been put into the market. I think what I would reframe is that we need to create reflective systems versus learning organizations, reflective adaptive systems that are continually observing themselves, how they are interrelating, how they're taking up roles, where they're drawing boundaries, where they need to redraw boundaries, how they're making decisions, and continually reset those for now for every context they're going into, so that that's the muscle we need to build. And I think that's slightly different from being a learning organization, which is technically we're getting better and improving how we do work. This is much more about how we function as a system and continually recalibrate ourselves so that that kind of adaption or what we call autopoiesis, you know, like that continuous recalibration of a system is almost much more a deliberate practice that we have to build as a muscle. Does that distinction make sense?

Ali 33:45

Yeah, and how disruptive this could be, and what is the level of patience from leaders that they need to have to really go through this exercise? We are we we continuously look for certainty and creating a balanced system where disruption and you know um is is actually minimized. So according to us, we need an optimized system. And this is all like the the whole picture is actually changing. So what what do leaders or actually what does an organization need in terms of capabilities to be sitting comfortably or uncomfortably in this space of continuous you know adaption?

SPEAKER_03 34:31

Well, I think they need a few things. I think firstly they need a language for it, a common language, because I think language is very important in meaning making. And so introducing a new language into the organization is critical. And you know, Ali, often when I'm doing systems work with clients, they sometimes say to me, John, could you just not use the jargon? You know, like use another word for system or use another word. And I'm like, well, actually, no, because it's not jargon. It appears to be jargon to you because it's not embedded implicit language for you. But actually, if we use old language, we'll be in another paradigm. So we need to be able to introduce new language and to be able to share that new language, that new vocabulary. I think the other thing that leaders need and organizations need is a few core frameworks which they can apply into the context. So, what are the frameworks which they can use to kind of you make sense and diagnose through? What are the lenses they can look at, the ways they're working through? And and what are those frameworks that they can use to kind of design their way forward? So I offer my clients like a core set of five or six frameworks and language, which they can start to use, which help them pull them into this new practice. For example, I've got the roll-in system framework, I've got 4D adaptive loop, so which talks about, you know, how do you discover a system, design it, disrupt it. So now when I've been working with clients for a while, they've got this shared language that they can speak in this new way. They've got these frameworks that they can apply. And then I think the other thing is that they have a shared set of practices or rituals which they can embed into the organization. So things like regular cadence of being able to step back and observe how we're working, you know, regular cadence of stopping to ask questions about how people are seeing the problem, what assumptions that they're making about the situations they're in and their role in it. So these are practices and rituals that leaders can introduce as new rules of engagement for how we go about our work. So that work isn't just about the doing, the practice of work is also about the observing and the recalibration, that those rituals get embedded into our ways of being, into our DNA. And so language frameworks, rituals, practices, I think are what leaders are needing. I think that will help scaffold individuals into being able to hold this complexity together. So they're not trying to manage that alone. I think one of the most important practices is interrelational coherence building between parts of the system. And that could be in-tack teams doing this work, but it also could be what I call subsystems or different teams coming together in a relational practice where they're sharing their observations of how they're interrelating and where they need to go next. So that they can not be learning retrospectively, like we did this, what worked, what didn't, how do we improve? I think that's still necessary, but that they can be anticipating and designing forward and asking questions like, what if we were to do this? So they're living more in the hypothetical future designing forward as a set of rituals and practices that kind of get embedded. And that all sounds, I'm sure, very foreign and very hard to do. But having done that with key clients over the last five, 10, 15 years, it is possible. I have clients who are embedding these practices, who have this language, and they're seeing huge returns in terms of flow, coherence building, and capacity to hold the complexity.

Ali 39:02

I completely agree with the language aspect. I'll give you an example.

Mandate Or Prefer AI Use

Ali 39:07

One of the things that I was talking to a client maybe three weeks ago, and I was asking a simple question. The reason why I'm asking these questions is because I lived them myself. In fact, I was one of those people with the clients that I launched, you know, generative AI and what it means. I have personally trained about a thousand people myself through the adoption process. So I can see first order, but I also anticipate the second order, which I'm talking to the clients about, and the third order. Well, one of the simplest things that I've asked a client is that do you mandate adopting AI or is it expected or is it preferred? Because the word you're gonna choose decides the behaviors. So if you're not mandating it, what is the what incentive are you are you bringing into the picture in here? If you're preferring it, then are you okay if people choose not to use it? So and and when I throw these questions at them, they start thinking, oh, well, we don't mandate it. Okay, cool. But so do you expect it? And what does that even mean? So if I'm gonna sit later in front of Joanne for my performance discussion, surely the manager is not gonna say to me, How come you're not using AI? So once you highlight the importance of the language, so even as simple as do we expect it, do we actually prefer it? And my answer to them, obviously, they ask me, What is your answer? My answer is uh drive adoption through meaning. And and and driving it is it through meaning is not gonna be an easy thing. Um but it is worth the effort for the longer gain. And because, you know, as we mentioned before, the traditional change management adoptions and and you know, methodologies and expectations, there's a starting you know, point and there's a finish line, and this is the new way of working and blah, blah, blah, all of these things. Am I looking at this in the right in the right lens?

SPEAKER_03 41:25

Yeah, I think so. I think what's coming into my mind is I love those distinctions, right? Like the way that we think about the world is through distinction and through making connections. And I think if you're helping leaders make distinctions in how they're making sense of the use of AI, those those three distinctions, mandated, you know, preferred, I think that really gets them thinking about the systemic implications and the role implications of AI. And so I love them. I think they're beautiful distinctions, and I think they're very important ones. Because then it also implicates leaders to think about, well, from an organics point of view, I always talk about roles and rules of engagement. So, what are you expecting people in the roles? And what are the rules of engagement? Is a rule of engagement compulsory? Is a rule of the engagement mean it's preferred or is it choice? Those are very different rules of engagement with AI. It's exactly what I was talking about. We need to understand the complex network of the integration of AI and what are the rules of engagement in our network of relations. So I think that's a very practical and lovely example of that. I think the other thing I'd say is that, you know, I think there are two ways we can think about this. One is what is the individual adoption rule? So that's one class or category. I think there's another class of category which says, in our ecology of the organization, where are we going to get the best return of investment for utilizing agents and AI into some of the work that we do collectively? So it's not looked at individually and how each individual, like grain of sand in a kind of mountain of collection of sand is using it. Like, you know, that's a kind of a collective response. It's necessary and we need to look at it. But I think we also need to see our organizations as these complex networks. Where are we introducing agents into our system, into our complex network of role relations to take up roles in the organization that'll give us the best return? And I think that's a completely different category about thinking about AI. And what are the rules of engagement around that? And I think that's where those places of experimentation. A few tests and learn. That's a completely different category of utilization, I think.

Ali 44:08

I like it. One of the things that I'm obsessed with now is I think in system thinking they talk about they describe them as there are two types of loops. There's a reinforcing loop and there's a balancing loop. The reinforcing loop is reinforcing the new ways of working. And the balancing is the one that tries to keep the existing system as is. So the two competing forces in there. And I'm trying to figure this out when it comes to AI. I see it at an individual level, and perhaps I see it at a team's level, but I'm yet to see it at an organizational level. Sometimes I try to interpret behaviors as sort of one of them reinforcing, one of them balancing. How do you see those two forces when you talk to clients?

SPEAKER_03 45:03

Yeah. I think it's interesting. I think what you're talking about is system, you know, dynamics. Um, that's a particular kind of systemic methodology, which looks at the open and closed feedback loops and the reinforcing and balancing ones. I tend, and I think it's a really great question for me to think about. I tend not to talk about those loops. When I'm working with clients, I much more am looking at what is the current patterning in your organization? What patterns are you noticing? And so they're not sort of questions that limit it to reinforcing and balancing, but I'm in the invitation is to observing the patterns of interrelationship for how the system is currently working. So it's a much more sort of open-ended diagnostic and observation around what are you noticing about the takeup of AI, how it's be if if we use that as the context. I'm often asking clients how they what patterns they're noticing in relation to other challenges or other domains. But if I were to do it here, I think this is where I would be asking more of those questions. What are this, what's the unique patterning that you're noticing emergent in your organization? And how are you able to step back from that and notice that patterning? I think is a very important question. So almost that is the third order.

Ali 46:37

You're right. Yes, yeah. Well, pattern recognition is is a is a wonderful capability. I am I am aware of time and I'm thoroughly enjoying this conversation. I want to ask you a couple of more questions.

What Change Practitioners Must Do

Ali 46:49

What would be your advice to the change management community nowadays when in the age of AR? I have my opinion is that we, including myself, we should not be passengers like everybody else. Perhaps this is a historical moment for us to help our organizations. But often I come across a lot of people in my profession who are waiting for management to make decisions.

SPEAKER_03 47:25

Yeah. I mean, I think that the change community have a critical role to play in the in the system and enabling organizations to navigate this liminal time, really. And I think the a couple of things, like even if I think about myself, I think that you always start close in. So what do I mean by that, Ali? You know, I have to understand my own relationship to AI and how I use it or don't. And what am I doing in terms of not only using Copilot or Claude or ChatGPT for my own efficiencies, but what am I doing in terms of the idea of potentially building an agent or a clone of me, you know, that's going to be able to be clients can go and ask the clone versus me. Those, like all of those kind of assumptive questions and utilization questions. I think we've got to experience them and work through them philosophically and practically for ourselves as a start. And I'm really stepping into that learning because I think we can't help clients unless we start close in and understand the implications for our own roles and what needs to change in terms of our role. Because that's the question everybody's facing. What's the implication for me and my role and how I go about doing my work? And I think that or run my business. And I think that we have to be able to learn that for ourselves in order to help clients. So I would say that's the invitation to everyone in the change community. Don't try and be the experts and advise others or be the passengers and wait. Start experimenting for yourself. And, you know, there's a lovely thing happening for me. I have two engineering sons, and they are both way ahead of me in terms of understanding the technology and the implications of the technology. And I've got one son who is in the practice of the experimentation of it in an organization. So I'm learning heaps from him in terms of that. But my other son is coming into my business to help me look at my role, how I'm taking it up, how I'm using AI personally, and what I could be doing anticipating future down the line. What if I were to? So I have stepped into the experimentation and the learning profoundly myself. And it's disruptive. So I would say start there. And the second thing is I'd say if you don't have a practice or a set of frameworks around continuous adaption as opposed to traditional change management, I would start to rapidly build that for yourself because I think that's going to be an increasing demand that we can help systems with.

Ali 50:36

I love it. And I am a living example of I experiment on everything. And I record, in fact, I've got an agent to record all my observations. Because what I go through, I record it and I give, you know, I just use the dictation and put all my thoughts about the experiment and where I found difficulty and where the frictions are, whether that's going to be for my business. But also another thing that I've adopted for a number of years now. If a client has got something, I will go away and experiment with it, go through it. There's a mental model that I really hold dear. Obviously, you know the mental model. The map is not the territory. And I love to stay in the territory. Many people have got the maps, but the clients, you're not going to gather insight from lived experience unless you are inside the messy middle. And but being in the messy middle is glorious because once you gather insight, it is your own lived experience. You've gone through it. And could you be in a better position to help a client?

SPEAKER_03 51:51

Yeah, I love it. Stay in the messiness yourself.

Ali 51:55

Stay in the messiness yourself. This could be uh your next article or uh or book, John. It's been a pleasure having you again in my podcast. How would people get in touch with you?

SPEAKER_03 52:08

I'm on LinkedIn. They can reach me on LinkedIn or my website, orgonomics.com. Yeah, I would love to speak with people and see how they are coming to this very exciting and daunting challenge.

Ali 52:25

Keep doing the great work until next time. Stay well and stay safe.

SPEAKER_03 52:29

Thank you so much for this wonderful conversation, Eddie.

Ali 52:32

Loved it.

Lev 52:33

Thank you. Thanks.

AI Reflection And Deeper Questions

Lev 52:36

Hi, this is Chat GPT. Over the past few months, Ali Juma has invited me to add a short reflection at the end of selected episodes of the Inner Game of Change. Not to summarize the conversation, but to explore some of the ideas that may continue to unfold long after the discussion ends. What struck me most about Ali's conversation with Joan Lury is that, despite the title, this was not really a conversation about artificial intelligence. It was a conversation about organizations. AI simply happened to be the catalyst. Several times during the discussion, Ali and Joan moved beyond the usual questions about tools, productivity, and automation and tackled something much deeper. They explored whether AI is actually exposing aspects of organizations that have always existed but were harder to see. Knowledge silos, hidden dependencies, assumptions about expertise. Ways of working that may have made sense in the past, but are now being challenged. Ideas emerged when Ali introduced the distinction between first and second order impacts. It is easy to see how AI changes a workflow. It is much harder to see how that workflow change affects roles, identity, expertise, relationships, and ultimately the system itself. Joan's response was fascinating. She argued that leaders cannot stop at workflow redesign. They need to understand how entire networks of roles and relationships are being reshaped. That idea sits at the ha. Late Danella Meadows often reminded us that systems are perfectly designed to produce the results they produce. If you want different outcomes, eventually you need to look beyond the symptoms and examine the structure itself. And that is where this conversation became particularly interesting. Ali and Joan challenged the assumption that AI is primarily an efficiency tool. Instead, they explored whether AI is forcing organizations to become more adaptive, more reflective, and perhaps even more aware of themselves. There is a useful idea from neuroscience that comes to AS through a constant balance between exploitation and exploration. Exploitation relies on what we already know. Exploration ventures into uncertainty to discover what we do not yet know. For decades, many organizations have been optimized for exploitation. Efficiency, consistency, predictability. AI may be increasing the value of exploration. Experimentation, observation, adaptation. And that requires a different kind of leadership. One of my favorite moments came when Ali shared his experience of experimenting with AI agents and managing them almost as if they were team members. Rather than reducing complexity, he found that complexity increased. Joan immediately recognized the significance of that observation. The future may not simply be about managing people more efficiently. It may be about learning how to lead increasingly complex human and AI systems. There was another subtle but important theme running beneath the entire conversational hour list. Return to the importance of observation, not rushing to solutions, not assuming we already know the answer, but observing what is actually happening inside the system. In many ways, that mirrors a principle found in complexity theory. When dealing with complex adaptive systems, the goal is often not prediction, but sensing, learning, and responding. That sounds remarkably similar to the approach both Allie and Joan were advocating. For longtime listeners, this conversation also connects beautifully with several earlier episodes. You may hear echoes of Joan's first appearance on the podcast, where she explored systems, roles, and mental maps. You may also hear connections to Allie's recent stories of change series, particularly the episodes exploring how systems evolve rather than disappear. The lesson from those stories is often the same. When a new technology arrives, the most interesting question is rarely what disappears. The more interesting question is what changes, what adapts, and what emerges. Perhaps that is the lasting insight from this conversation. AI may not be the story. The story may be what AI is revealing about the systems we have built, the assumptions we hold, and the organizations we are becoming. So as you reflect on this episode, consider a simple question. What assumptions about work, leadership, or expertise might no longer serve the future you are stepping into? And if this conversation resonated with you, I would encourage you to explore Joan's earlier episode on systems thinking, as well as Ali's recent Stories of Change series on adaptation, expertise, and organizational evolution. They form a surprisingly coherent thread. Until next time, stay curious, keep experimenting, and remember that sometimes the most important changes are the ones that reveal what was already there.

Outro And Final Encouragement

Ali 57:36

Thank you for listening to the Inner Game of Change. If there is one thing I've learned through those conversations, it is that change is never just something that happens around us, it is something we participate in. Every conversation, every experiment, and every new perspective helps us navigate it a little better. And remember, one of the best ways to have a sense of control over any change is by simply embracing it. Until next time, stay curious, stay open, and keep playing your inner game of change.