The Leadership Pipeline Isn’t Broken. We Deleted It.
We Deleted the Classroom and the Teacher. Now What?
The dashboards look clean.
A lot of organisations spent the last three years making structural decisions that are now reflected in the numbers: leaner teams, reduced headcount, faster decision cycles, lower overhead. Middle management layers have been compressed. Reporting functions have been automated. The efficiency ratios that used to keep CFOs awake now look manageable. The strategy is working.
I want to talk about what’s underneath the dashboard.
Yesterday morning on RTHK Radio 3’s The Brew, Phil Whelan asked me to break down what’s happening with the leadership pipeline. He framed it exactly right: automating entry-level tasks and cutting middle management is quietly wiping out the corporate learning ladder. If AI takes over all the junior roles, nobody actually learns how to do the senior ones. You end up with brilliant algorithms and future bosses who haven’t got a clue how anything works.
That’s not a colourful hypothetical. It’s a description of something already forming.
The Reporting Illusion
For roughly a decade, the dominant critique of middle management centred on its relationship with reporting. Middle managers, the argument ran, spent too much time aggregating information upward, summarising what their teams were doing so that senior leaders could see across the organisation without talking to everyone individually. This was expensive, slow, and structurally redundant. AI could do it better, faster, and without the salary.
The critique wasn’t wrong. Reporting is precisely the kind of pattern recognition and synthesis task that AI performs well. The savings are real. The efficiency argument was legitimate.
But the critique confused the visible function with the actual function.
A middle manager’s job was never reporting. Reporting was just the part of the role you could measure, put in a performance review, and therefore most easily argue was replaceable. The actual job was something that doesn’t appear neatly in any job description: the hallway conversation with someone who was struggling before it showed in the data; the translation of an executive strategy into something a team could actually act on given the conditions they were operating in, not the conditions the executive suite imagined; the deliberate assignment of increasingly consequential work to someone who had potential, precisely because that person needed to learn by doing; the holding of organisational culture in the gaps between announcements and all-hands meetings.
AI produces a report. It can’t have the conversation that changes someone’s career.
The DDI Global Leadership Forecast found that 71% of leaders report high stress from the pace of organisational change, and (this is the number worth sitting with) 40% of those stressed leaders are actively considering leaving. Not disengaged. Not quietly coasting. Actively considering exit, from the people who are still in the roles.
When you read that alongside Gartner’s research showing 20% of organisations have already committed to eliminating 50% or more of their remaining middle management roles by end of year, you get a picture of a pipeline that’s been cut and is now losing pressure from what’s left.
The people who survived the cuts are leaving because the role has been hollowed into something that no longer resembles what they were trained for, selected for, or understood themselves to be doing.
The Other Deletion
While organisations were compressing the middle, something structurally similar was happening at the entry point.
AI was taking the junior work.
Not uniformly, and not completely. But across wide sections of professional life (legal, financial, consulting, marketing, technology, research, media, analytics) the low-stakes, consequential work that used to be how entry-level professionals built judgment has been automated, reduced, or redesigned out of existence.
I wrote about this earlier this year in We Didn’t Just Automate the Jobs. We Removed the Years. The entry-level job was never really a job. It was a laboratory. It was the place where someone who knew nothing about professional life learned the hidden curriculum: how to sit in a meeting room and read what wasn’t being said; how to handle a difficult brief from someone who expected better; how to recover from a mistake without the floor falling out from underneath you; how to build a sense of what matters and what doesn’t. The professional judgment that can’t be taught in a classroom and doesn’t appear on any syllabus.
That laboratory is being quietly closed. Not because organisations made a deliberate decision that professional development doesn’t matter. Because the decision to use AI to perform entry-level analytical and administrative tasks made economic sense at the individual level, project by project, team by team, without anyone asking what the cumulative effect on capability and succession would be.
Forbes named the consequence directly in June this year: “Management Cuts Today Are Shaping A Leadership Shortage.” Not might shape. Are shaping. Present tense, already in formation.
So here’s where we’ve arrived. The pipeline has been cut at the bottom and in the middle at the same time. Entry-level roles have been automated or redesigned to remove the developmental component. The middle management layer that was supposed to receive those entry-level employees, develop them through years of increasingly complex work, and eventually produce senior leaders from them has been significantly reduced.
Nobody ran the numbers on what that produces in five years.
The Classroom Question
When Phil pushed on this on air, the question he returned to was: where do people learn now?
It’s the right question. The traditional path was legible. Entry-level job: make mistakes under supervision, build judgment through consequential low-stakes work, get better, get promoted. Work under an experienced manager: watch them make the hard calls, understand why they made them, get stretch assignments that accelerate your development, eventually become that person. The path was long (sometimes frustratingly so) but it was a path. It had structure, sequence, and accumulated logic.
That sequence has been interrupted at both ends simultaneously.
The classroom (the entry-level role as developmental space) has been deleted.
The teacher (the middle manager with the contextual intelligence, institutional memory, and deliberate coaching practice) has also been deleted.
Phil’s question is unanswered for most organisations: what replaces both?
I’ve been working in organisations across Australia, Asia Pacific, and internationally for more than 30 years, and the answer I’m getting from most of them right now is: nothing yet. There are individual organisations doing deliberate, thoughtful work to redesign their professional development pathways for a post-automation environment. They exist, and they’re building something genuinely different. But they’re the exception.
Most organisations have made the efficiency decision and deferred the capability question. The savings are in this quarter’s results. The capability consequences are in 2028’s problem.
That’s not a sustainable position.
The Succession Void That’s Forming
The DDI data contains a third-order consequence that isn’t getting nearly enough attention.
The 40% of remaining leaders actively considering leaving aren’t a retention statistic. They’re the beginning of a succession void.
The people who were supposed to become the next generation of senior leaders, the ones now in middle roles who should have been developing the cohort below them, haven’t had the developmental conditions to do that developing. The entry-level pipeline that should have been feeding capable, experienced early-career professionals into their teams has thinned. And now, the managers themselves are leaving.
What they take with them is very difficult to reconstruct quickly.
Contextual intelligence: the ability to read an organisation, understand its informal power structures, know which decisions are actually decisions and which are already made. Institutional memory: what was tried in 2019 and didn’t work, and why. Relational authority: the trust built through years of visible, consistent, human presence in the organisation. Organisational fluency: the ability to navigate the human complexity that sits underneath any strategic initiative without having to be taught from scratch.
You can’t hire this at scale from the outside. You build it over years, through exactly the developmental conditions that have now been removed.
The 2029 question, and I say 2029 because that’s when the compounding effects of three to four years of thinned development pipelines start to crystallise, is: who makes the high-stakes judgment calls? Who has the contextual intelligence that AI-augmented decision-making requires a human to provide? Who carries the institutional knowledge that allows an organisation to function under pressure rather than just under normal conditions?
Which organisations can answer that question confidently?
In Australia, the picture is sobering. Microsoft’s 2026 Work Trend Index found that only 28% of Australian businesses have an aligned AI and human strategy. The other 72% are operating in a gap: AI capability moving faster than the human infrastructure designed to work alongside it.
That’s not a technology problem. That’s a design gap, with real consequences arriving on a delay.
Efficient Organisations and Inhabitable Organisations Are Different Things
I’ve been drawing this distinction for several years, and it’s never felt more applicable than it does to this moment.
An efficient organisation has optimised its costs, its structures, its processes. It knows where every dollar goes. Every role can be justified in terms of a measurable output. The overhead is lean and deliberate. It looks excellent on paper.
An inhabitable organisation is one where people can actually live, trust, work, and develop over time. It has space for mistakes to happen and be recovered from. It has the relational infrastructure that holds people through uncertainty. It has the developmental conditions, formal and informal, that allow professional capability to be built over time rather than imported on demand.
These two things are not automatically the same. They don’t have the same durability under pressure.
An efficient organisation that has removed its developmental infrastructure feels excellent right now. The numbers reflect genuine improvement. The dashboards are cleaner. But underneath the efficiency, the capacity to develop leaders, absorb organisational anxiety during periods of change, govern AI-augmented decisions well, and regenerate capability after people leave has been quietly removed. The system looks healthy. It’s hollow in a specific place.
The confidence-preparedness gap I wrote about in June sits here. Leaders who believe they’re prepared for disruption because they’ve optimised their structures, but who haven’t asked whether the human capability to act on that optimisation is still in place.
The organisations that feel most efficient right now may be the ones least prepared for 2029.
For Young Professionals: It’s a Structural Problem, Not a Motivational One
Phil raised the HK angle on air, and it applies equally across Australia and the Asia Pacific: highly educated young professionals entering a professional market where the rungs of the ladder have been removed.
The entry-level roles where they expected to cut their teeth: reduced or redesigned to extract the output without the development. The experienced managers they expected to learn from: also reduced, or stretched so thin across remaining teams that deliberate mentorship has become impossible rather than merely infrequent.
The young professional in 2026 is trying to build a career inside a structure that’s been compressed at both ends. Their response to that: frustration, disorientation, accelerated job hopping in search of somewhere that does invest in them, is being misread as a generational attitude problem.
It’s not. It’s a rational response to a structural condition.
And the organisations that understand this, that the young professional’s experience of a hollowed learning environment is accurate, not a matter of attitude, are the ones making deliberate choices to create structured developmental pathways: formal mentorship with real access and accountability, deliberate project design that gives early-career people consequential work under guidance, and explicit succession architecture that isn’t left to chance.
What to Do With This
This isn’t abstract. These decisions are sitting in front of most organisations right now.
If you’re still working through structural decisions about middle management, the question worth spending time on is not “can AI do the reporting function?” It can. The question is: “What happens to the developmental function that was embedded in that role, and how do we ensure it survives the restructure in deliberate form?”
If you’re looking at graduate and early-career hiring and asking whether AI can do the work those roles were doing, it probably can do the outputs. The question is: “What replaces the developmental laboratory those roles provided? Are we building that deliberately, or are we assuming the market will supply senior capability when we need it?”
If you’re in a middle management role and you’re one of the 40% thinking about leaving, the organisations that will compete well over the next five years are the ones that understand what the role actually does. They exist. Finding them is worth the time.
And if you’re a senior leader looking at your succession depth right now and feeling uncertain about it, that feeling is data. It’s worth acting on before the void becomes undeniable.
In 2019, when I argued that Human First had to mean something structural, not sentimental, not just a values statement, but an actual design principle for how organisations use technology alongside people, the organisations that heard it most clearly were the ones that refused the false trade-off between efficiency and humanity. They pursued both. They used technology to do what technology does well, and they invested deliberately in what only humans do well.
That same design principle applies now, to a more specific and urgent version of the same question. AI can do the junior analytical work. What replaces the developmental laboratory? AI can produce the management report. What replaces the coaching conversation, the culture-holding, the deliberate development of the next generation?
These aren’t rhetorical questions. They’re design questions. They have answers. They require deliberate attention, an explicit decision, and someone accountable for executing it.
The double deletion isn’t irreversible. But the window to redesign before the consequences arrive at scale isn’t open indefinitely.
Morris Misel is a foresight strategist and keynote speaker based in Melbourne, working with organisations, associations, boards, and media on preparing for uncertainty and making better strategic choices. This post draws on a conversation with Phil Whelan on RTHK Radio 3’s The Brew, 28 July 2026.
Related reading: We Didn’t Just Automate the Jobs. We Removed the Years | In 2019 I Said Human First. Organisations Heard ‘Efficient Last.’
Choose Forward.
▶ Full transcript , RTHK Radio 3 The Brew with Morris Misel and Phil Whelan, 28 July 2026 (16 min)
on Radio 3. So let’s whiz over to St. Kilmer and Melbourne, the lair of very vlogineas. Morris, Miss Alowski, how are you? I’m excellent, thanks. Happy Tuesday and glad to have you back. Happy Tuesday. It’s good to be back. I pranged myself. We’ll do it now and again. So basically, Morris has written a couple of very cool articles, details of course, on the bruise Facebook page. But as I said, putting it simply, and I may be wrong, but I only had five if I take over the junior roles, nobody actually learns how to do the bigger roles.
So we’ve got brilliant science down there, future bosses who haven’t actually got a clue. This does sound quite logical. It comes a bit more. Yeah, well, they’re pretty much set up. We’re done for the afternoon. Yes, absolutely. You’re absolutely right. I mean, this is one of the things I’m grappling with, and I’ve been saying to a lot of people over the last year or so, I don’t think we’ve quite realised, and I’m talking to my corporations more than you and I in the audience, yeah. They’re getting really excited about our official intelligence and what it can do. And of course, what it can do really well is the routine work.
Well, really simple. If it’s got something to do, it’s got somebody to hold its hand and make sure it keeps doing it the way you want it to. Little details. Yeah, exactly. So if you take the lawyers, for instance, you know, take the legal. What you had these young punks come in, you know, they’ve done their five, six years at uni, they’ve come in. What you get them to is really the grunt work. They sit there and they go through the archives, they go through all of the books, they write the briefs, they do all this work. They’ll be at the end of the first class honours. Yeah. Yeah. Yeah. And they go and get the coffee and the laundry and all that kind of stuff.
Everything we’ve seen in the books, everything we’ve seen in the movies. Yeah, yeah. Now, apart from the laundry and the rest of it, which maybe AI can do and bottom or whatever, a lot of that stuff is now better done, quicker done, cheaper done by AI and technology. So the question is, and it’s increasingly not just the question, it’s increasingly showing up now, when I look at the internships, when I look at the companies and who they’re not employing at the end of each year, which is traditionally when they bring in next year. So the end of the year means the graduates are finished in our case, they’re finished around November and you get a whole slot of new graduates coming into the firms around January, February.
And they start their long careers of 40 years or so. Yeah. Do they still do that? No, they don’t. They don’t. They don’t. They don’t. I’m exaggerating. But that’s when they start their life. Now, what’s not happening is that a lot of those junior people are getting jobs because there isn’t a job for them. So my question has been for, you know, for the last couple of years, increasingly loudly, either the last couple of months, is if you’re all your companies who are putting AI in there and things that are doing great work and you are, you’ve really got to think about, that’s a great thing for today, but you’re costing yourself, because in four, five years, the people that would have naturally come up through the ranks, that would have learned on the job, maintenance stakes under the tutelage of other people would have learned by hard knocks and just seen how other people do the work. It doesn’t matter what the work is and then been ready to take
on more responsibility. Where are those people coming from? More and more, Maris. It does seem, I mean, in this realm of doing those kinds of jobs, it does seem like AI is like a wacky teenager that like sugar buses every now and again. It does need oversight. And we’re not really talking about that. Are we? No, we’re not, but we should. I mean, yes, I’ve accepted all that and saying a company needs to do that, but my question is building that film. I mean, what happens to people and most of us cut our teeth on the entry level jobs. It makes sense if you’re young, 17, 18, 24, whatever it is when you finish your apprenticeship or you finish your uni or you’re just going straight out of the school into a job. It doesn’t matter. None of us, as much as we want to, we’re starting at the top. Most of us are starting at the bottom. That’s not a negative. That’s the way the world of work works. The happy medium, Maris seems to be that this is used to pay unless. Well, there’s not even an excuse to pay, Maris, because they’re just not getting the job.
And I’m not even working about what we’re talking about. Well, we’re talking about legal, we’re talking about accounting, we’re talking about engineering, we’re talking about apprenticeships of all kinds. It’s really anywhere, even the company, the small one, two, three, four person company who might have hired a gofer or a young person to do things. A lot of that’s not necessary now because you can get AI, you can get machine, you can get all kinds of things to do it. Now, of course, this is not 100% of every job in the world, but a lot of jobs increasingly are going. A lot of tasks and a lot of these entry level positions will become increasingly difficult to find. I’m talking about somebody looking for them. And companies will just not be sourcing them why would you pay somebody in Australia in the nakula? Yeah. $50, $60, $70,000 when they’re out of uni to do this kind of grunt work before you did because you could farm it out to a client. They paid for it. There was a whole pyramid of systems of the way these organizations worked.
Basically, the grunt work was done, the majority grunt hard work was done by the cheapest person possible and it then went up through the layers and the client must probably saw somebody at the management or higher level depending on what they were going to pay. But when the grunt work person makes a mistake it tends to be just a human type, you know, cock up. But when AI makes a mistake doing those kind of jobs it’s telling your chalk is in fact cheese. Yeah. And the difference is when the when the when a human when you and I make that kind of mean, not you, because you don’t make mistakes, but when I make a mistake at those junior levels somebody comes over and of course you’re not allowed to. I’m just being euphemistic here they hit you over the head of course you’re not allowed to but somebody pulls you up which is what you need and you figure out quite quickly, you know, it might be terrible to your self-esteem, whatever happens, but you know that something has not worked the way it should. Somebody’s explained to you hopefully why not and what could have been better or done differently and you’ve learned
the next time you do it differently. There’s a sidebar question for you. So at this stage of the game with AI, we’ll stick with the clerical stuff. Do you think the company’s and the brains behind AI are getting closer to it’s to stopping it being cranky and losing its memory because that is the biggest, those are the biggest criticisms on this level we’re getting. So losing the memory absolutely. So that’s almost that’s almost baked in now. If you’re looking at claw and chat GPT their memories, their ability to keep memories much better, built into the back end of a lot of clawed now is the ability for it to go from one session to another whereas before you write it used to cold stop and cold start. Basically I had to teach it again like a goldfish. You know, I’ve finished a half an hour. I don’t care. I’m a goldfish. I’ve forgotten everything in the last half hour start again. That’s not as true as it once was and there are very, very good work arounds for that now. But lots of people use their documents. There are all sorts of things you can do. So that’s side of it’s getting better. There is one weird little bit of crankiness to do
with this one. Gemini CEO recently said we have some problems. Our connectivity to our own apps is not cooking. It’s the single is the first and second biggest complaints one or the other from user saying your sandbox issues are insane. And he admitted it and he said we must do better. And that’s Google man. Yeah. And they are doing better. So he’s absolutely right. But he’s talking about the commercial product that you and I can use right now. He’s not telling you in fact what you know what they’re doing 18, 24 months. No, no, no, no. He’s talking about now the one that people say that’s what people pay subs for and I think they have enough. Oh, they have and they rightly have. But again, it’s much better than it was two years ago. I’m not arguing for it.
Yeah. And I do know some of what’s coming ahead and some of what’s coming ahead is extraordinary in the way of what it will be able to do the way it will capture everything that you’re rightly are saying now that loses. That because it’s a known problem, it’s kind of like the battery on the phone. Everybody knew it was wrong and bad. It just took a long while to figure out how to make a battery on the phone last longer. It’s pretty much where they’re at now. It’s a known problem. Everyone wins just in the plane. It’s rightly about it. Absolutely right. Exactly. And they’re working on it really hard to fix it. It’s quite funny actually because they’re this company that they’ve really led the way in so many areas. But when things go wrong, they don’t jump out and say, hey, we’ve got a problem here and we’re fixing it. They tend to.
And then you say, well, why isn’t sort of Google coming forward and saying, what’s going to? Well, it tends to be a drift out, a rolled out kind of thing and companies don’t do this. All sorts of excuses. I’ve said to you many times that I think that a lot of what’s being called AI now is an excuse to do things in a corporation. So we’re letting people go because we can employ these machines. I’m absolutely confident, roll gold confident that people will start being employed again once companies realize that these AI and machines are great and terrific. And they advocate for them in the right uses and the right places aren’t and everything panacea that you need people. We’re going to start pushing back this to me. AI is being used as an excuse to do. And we’re in the honeymoon, aren’t we? Well, we are, but it’s bigger than that because it’s we also are living in a very turbulent economic time. We’re building, you know, socially, it’s very difficult. They’re all sorts of things happening in the world that make it not as viable to keep staff on. And again, through COVID, a lot of companies really put on a lot of staff coping or thinking
they were going to cope with what these people could do and would need them. And now, and I’m talking about the very large companies, especially the Google Facebook, all of those are basically admitted. They have too many people and it was based on an old model. So what they’re doing now, they don’t say this is his mind’s reputation is that they have used artificial intelligence as the reason to get rid of a whole lot of people. We don’t need them anymore because AI is doing it. Let me refer our listeners to one of your articles here, Morris. Yeah. Yeah.
Obviously, as always, Morris Futures.com. So he says in 2019, I said human first organizations heard efficient last, we didn’t obviously not be to teach these junior guys and girls how to gate keep these various models. Yeah, it could be, but you know, I just get to teach them anything. I’m being stupid. But they’re probably brilliant. They don’t need teaching. I mean, they are.
And their job is not to keep the AI. In fact, often with junior people, they don’t even have access to artificial intelligence in the way that we think they might. And also, more importantly, nearly every organization I talk to, they don’t have the authority. They’re basically being told, this is what you use, this is how you use it, this is when and where you use it. And don’t forget loads of corporations, they’re barking mad about security and very often it’s security.
So they can’t be. Do it. So they can’t be with their equipment to facilitate the use of this wonderful stuff. And a lot of the people that are making these decisions really aren’t very sure about what they’re making decisions about or on. And that’s not saying they’re not clever. It’s saying that this thing is too new, it’s moving at a pace, it’s really difficult to get a handle on it.
And these people are not people that are skilled at looking at these things. They didn’t have to be. This was not part of their normal work understanding up until literally two years ago. So there’s that issue and the other one I pointed out in the article, which is kind of we’ve alluded to, this is my double whammy, that because they think that middle management, which is basically managing people and managing people means there’s a routine so we can understand what they’re doing, when they’re doing what the output is, all those kind of known outcomes. Because that’s what middle management is often seen as, many corporations have got middle management or they’re getting less of them.
The middle management is, you know, the middle person is, whatever we’re saying that will correct word is. But we’re saying that a lot of people statistically, and this was a piece of research that came out, only in the last couple of months from DDI global leadership. And they are a fairly well known company in this space, we’ve been doing research for quite a while. They said that a vast majority of companies are talking about getting rid of middle management. And they put some percentages behind that.
But what they’ve talked about is that the notion of middle management is disappearing. I’m saying the notion of entry level jobs are disappearing. Well, who’s left? So the blue who you said will be unskilled, but they’ve got the top jumps. So you’ve got the people who have been there forever, who know the work really, really well, who don’t necessarily, nor should they want to do the grunt work. Yeah, who don’t know AI really well enough to be able to instruct that nor do they want to.
They are working with people who are kind of in the middle of their, you know, in the middle of learning, but haven’t quite learned, haven’t been taught the way they want to. So I think you’re ending up with this real unusual structure of corporations moving forward. Are you based on Australia mostly? Well, you’re your manor sort of thing. I am based in Australia, but the stats I’m pulling up are from all over the place in Scotland, which is another analysis company. Yeah.
Already said that they have committed, so this is research across their clients across the globe. Mm-hmm. But 20% of the organizations that they have research, this was July this year, have said that they are getting rid of 50% of their middle management roles. Yeah. So 20% of the people they have interviewed, and in corporations, have said 50% of middle management is going disappeared by the end of this year.
So what we’re talking about colors, isn’t it? We have to remember horrible humans. What where do they go? What are they going to do? Thank you. And that’s my issue too. I mean, what do we do with these people? I don’t think that we’re wholesale losing jobs.
I think this is just a real structural period. We’re in a consulting word, we’re in the yuck period. It’s just terrible. And I say that because I know every one of that is attached to a human being with a life and the family and all those things. I accept all of that. We’re in this terrible period of uncertainty and turbulence, not just because of AR, but because of what’s happening around the world and the uncertainties and business and all kinds of things.
And if we’re losing all these entry level jobs, which I’m arguing we are, and middle management jobs are disappearing because we know they are. Yeah. Then we have a whole new workspace that we need to recreate. It’s not the first time, is it though? Let’s go back to about 17. No. The industrial revolution and then steam took over everybody’s job and if we went.
absolutely. But even more recently, the internet, the PC, the PC was a great one for that. We talked about how many people would lose their jobs because everyone would have a PC on their desk, the internet did it, social media. I mean, I can give you a list of jobs. I mean, we should do that as a thing one. There’s a job that no longer exists. And this is where we’re at now.
The difference now is literally the speed that we’re doing it. We’re collapsing these businesses. I mean, the disappearing, we’re collapsing the businesses and what’s happening at the pace. And the world around it is also moving at the pace. Is this a good time or a bad time to be alive? Do or does it depend on you or? I think you’re age group. Look up.
I think it’s always a great time to be alive. The alternative is how good it does help. We’ve got about a minute and a half more. I’d like you if you would just to take me through the other article, which I should have spent more time on. We didn’t just automate the jobs we removed the years. Give me the A, B and C. Both of these are on his website.
We’ve pretty much already talked about that. So that was where I thought that what we’ve done is we’ve taken out that early years where people learn their jobs And that’s where we’re moving the years and also because we’ve got rid of manual management We’ve removed the years from there as well So we have these people that between entry between middle management and then above middle management So those two rungs are gone. So those middle years where a company does a lot of its work and gets a lot of its activity from Don’t exist in the way they did before. Are all these job stratters and titles and Levels are they actually based on government going way back or is it something else?
No, it’s actually based on industrial revolution model of how to get worked down efficiently. Okay Yeah, I suppose in those days it did Well, it did make perfect sense because what you did was put together a whole lot of people into an artificial environment In other words into a factory into an office Yeah, and you put somebody in charge of them somebody in charge I had to know what everybody was doing and that made perfect sense because that meant you could skill Education system work towards that culture work towards attention taxes everything work towards one person one task You know one set for life and now all of a sudden we’re saying well that doesn’t work anymore It’s broken there is always one thing that happens when the problem is this happened Morrison that is I’m assuming that the death rate of a generation nice happy ending goes up because they fall into machines and stuff that they weren’t aware of before Now, I think we’re all gonna survive it will be good on all shorts and there is not our next president We just need to get through this thing
Take care Moris brilliant to talk to you as always I catch you next Tuesday. M that Morris Futures.com go to the blog always
Frequently Asked Questions
What happens to leadership development when middle management is eliminated?
When middle management is reduced significantly, the informal coaching, culture-holding, and deliberate talent development that those roles performed doesn’t disappear automatically: it goes undone. The DDI Global Leadership Forecast found that 71% of leaders report high stress from the pace of change, and 40% are actively considering leaving. This is partly because the roles have been stripped of the human functions that gave them meaning and made them effective. Without that layer, organisations lose the translators between strategy and execution, and the developers of the next generation of leaders.
Why does AI removing entry-level jobs create a succession problem?
Entry-level roles weren’t just about producing outputs. They were the developmental laboratory where early-career professionals built professional judgment, the hidden curriculum of how to read a room, navigate complexity, recover from mistakes, and build institutional knowledge. When AI performs those outputs and the entry-level role shrinks or disappears, that developmental experience disappears with it. The succession problem arrives years later, when the people who were supposed to become senior leaders haven’t had the formative conditions to get there.
What is the “double compression” in leadership pipelines?
The double compression refers to organisations cutting the pipeline at both ends in the same window: removing or automating entry-level roles (where people learned), and simultaneously eliminating significant portions of middle management (where people were supposed to be developed and who were supposed to do the developing). The cumulative effect is a leadership pipeline that has been thinned at the point where capability enters and at the point where it was supposed to be grown. Nobody modelled the combined effect on succession depth.
How does Australia compare on AI and human strategy alignment?
Microsoft’s 2026 Work Trend Index found that only 28% of Australian businesses have an aligned AI and human strategy, meaning explicit, documented alignment between what AI does and what people do in their organisations. The remaining 72% are operating without that alignment, allowing AI adoption to outpace the human infrastructure designed to work alongside it. This makes Australia particularly exposed to the consequences of the double compression, because most organisations haven’t designed a response to it.
What distinguishes efficient organisations from inhabitable ones?
An efficient organisation has optimised costs, structures, and processes, every role justified by measurable output, overhead lean and deliberate. An inhabitable organisation is one where people can live, trust, develop, and thrive. It has space for mistakes and recovery, relational infrastructure, and deliberate developmental conditions. These two things aren’t automatically the same. An organisation can be highly efficient while having quietly removed the capacity to develop leaders, absorb organisational anxiety, and regenerate capability after people leave. Inhabitable Futures, futures people can actually function inside, require deliberate design, not just efficiency optimisation.
Morris Misel is a foresight strategist and keynote speaker based in Melbourne, working with organisations, associations, boards, and media on preparing for uncertainty and making better strategic choices. He appears regularly on RTHK Radio 3’s The Brew. This post draws on a conversation with Phil Whelan, 28 July 2026.
Related reading
We Didn’t Just Automate the Jobs. We Removed the Years.
In 2019 I Said Human First. Organisations Heard ‘Efficient Last.’
