Are Managers Becoming Obsolete or More Important Than Ever in 2026

The rise of AI and flatter organisational structures has sparked intense debate about the future of managers, according to Fast Company, which gathered eight industry experts on whether middle management is endangered or evolving. The sharpest answer is that AI is not making managers obsolete, it is making one specific type obsolete, the one whose entire value was controlling information flow. At Metaintro, we track roughly 50 million job postings, and this is a crucial distinction for anyone weighing a management track. Some management skills are dying while others are becoming more valuable, and knowing which is which is how you decide where to aim.

Which Kind of Manager Is Becoming Obsolete?

The information router, the manager whose core function was moving and controlling information.

The historical role is clear. For decades, middle management existed because information was scarce and hierarchy was the routing system, with managers moving data upward, translating strategy downward, and approving decisions in between. When information was hard to move, humans were the network that moved it.

AI collapses that function. The expert view is blunt, that AI collapses this function quickly, and it should. When information flows freely and systems can route and surface it automatically, the manager whose value was being that routing layer loses their core purpose.

The obsolete manager is specifically defined. It is the one whose entire value was controlling information flow. If a manager's main contribution is being the gatekeeper who information passes through, approving routine things and relaying messages, that contribution is exactly what AI now does faster and more reliably.

The lesson for anyone in or considering management is direct. A management role built primarily on controlling information, approvals and reporting is the vulnerable kind. Recognising whether your management value is mostly routing, or something more, tells you how exposed you are, much as the manager-support gap exposes teams whose leaders were never trained to develop them.

What Kind of Manager Is Becoming More Valuable?

The one who governs the intelligence layer, which the experts describe as a harder job, not a simpler one.

The replacement is more demanding, not less. The key insight is that what replaces the information router is not a simpler structure but a more demanding one. AI-native organisations do not just have fewer managers, they need managers who can design and govern the intelligence layer itself.

The new questions are the hard ones. The role centres on questions like who owns the escalation thresholds, who decides which decisions stay with the system and which require human judgment, and who ensures accountability does not disappear when a workflow becomes automated. Those are not lighter responsibilities, they are heavier ones.

The concept has a name. One expert calls it decision architecture, and describes an engagement where removing manual approval layers did not simplify the management structure but raised the skill floor, because the team suddenly needed people who could interpret what the system was doing, intervene at the right threshold, and redesign the workflow when conditions changed.

The distinction is the whole point. The managers who remained were not the ones who knew the process, they were the ones who understood the logic behind it. Knowing how to execute a process is replaceable, understanding why it works and how to govern it when a machine runs it is not, which overlaps with what gets people promoted now.

Why Do Flatter Structures Not Mean Fewer Skills?

Because removing layers raises the demands on the managers who remain, rather than lowering them.

The intuition is wrong. It is tempting to assume that flatter organisations and automation mean management gets easier or less necessary. The expert evidence points the other way, that removing approval layers raised the skill floor rather than lowering it.

The reason is that automation shifts the work from doing to governing. When manual approval layers are removed, the remaining managers need to interpret system behaviour, intervene at the right moments, and redesign workflows as conditions change. That is more cognitively demanding than approving items one by one.

The failure mode is instructive. The organisations that struggle are not those that eliminated management layers, but those that automated execution without building governance around it, which the expert calls a management design failure rather than a technology failure. Removing managers without replacing them with governance is where things break.

For a worker, this reframes flat organisations entirely. A flatter structure does not mean management skills matter less, it means a different and harder set of management skills matters more. The people who thrive are those who can provide the governance that automation requires, not those who simply had a title in a hierarchy.

Should You Still Pursue a Management Track?

Yes, but pursue the right kind of management, because the answer depends entirely on which capabilities you build.

Avoid building your value on information control. If the obsolete manager is the information router, then a management path centred on approvals, reporting and being a communication gatekeeper is building toward a role that is disappearing. That is the version of management to avoid making your core identity.

Aim for judgment and governance instead. The valuable manager understands the logic behind the process and can govern automated systems. Developing judgment, the ability to decide what needs human input, to set the right thresholds, and to redesign how work flows, is building toward the role that is becoming more valuable, and it is closely tied to the AI skills employers actually want.

Develop the specific capabilities the experts name. Owning escalation thresholds, deciding which decisions stay with the system, and ensuring accountability survives automation are concrete, learnable skills. Deliberately building them positions you for the demanding, durable version of management rather than the fragile one.

The honest framing is that management is not dying, it is bifurcating. One branch, the routing and control branch, is shrinking. The other, the judgment and governance branch, is growing and getting harder. Choosing to develop toward the second branch is how you make a management track a good bet rather than a risky one.

What Happens to Organizations That Get This Wrong?

The experts are clear that the danger is not automation itself but automating without governance, and understanding the failure mode helps you spot healthy and unhealthy employers.

The failure is removing people without replacing the judgment. The struggling organisations are those that automated execution without building governance around it, which is described as a management design failure rather than a technology failure. When a company strips out managers to look lean but leaves no one accountable for how the automated systems behave, problems accumulate unseen.

The symptom is accountability disappearing. A central concern is ensuring accountability does not disappear when a workflow becomes automated. In poorly designed automation, decisions get made by systems that no human is clearly responsible for, so when something goes wrong there is no one who understood the logic well enough to catch or fix it.

The contrast with healthy design is instructive. The organisations that do this well do not just have fewer managers, they have managers who design and govern the intelligence layer. The difference between a company that thrives on automation and one that struggles is whether it invested in governance or just cut headcount.

For a job seeker, this doubles as a way to evaluate employers. A company automating thoughtfully, with clear ownership of thresholds and accountability, is a healthier place to build a career than one cutting management to look efficient while leaving its automated systems ungoverned. Asking how an employer handles that governance tells you a lot about how well run it actually is.

How Do You Build Governance Skills Now?

You can start developing decision-architecture capabilities in almost any role, well before you hold the title.

Practise understanding logic, not just process. The distinction the experts draw is between knowing the process and understanding the logic behind it. In your current work, make a habit of asking why a process exists and how it would need to change under different conditions, rather than only learning the steps. That shift from execution to understanding is the foundation of the durable skill.

Learn how automated systems in your work actually behave. As AI enters your workflows, treat understanding its behaviour, where it is reliable, where it fails, and when a human should intervene, as a core skill, the same discipline behind managing AI agents without burning out. The people who can interpret and govern these systems are exactly the managers of the future the experts describe.

Take ownership of thresholds and judgment calls. Even without a management title, you can practise deciding when something needs escalation, where the line between automated and human judgment should sit, and how to keep accountability clear. These are the specific responsibilities the new manager owns, and rehearsing them builds the capability early.

Think in terms of workflow design. The valuable manager can redesign the workflow when operating conditions change. Developing the habit of seeing not just your tasks but the system they sit within, and how you would improve it, is how you build the design mindset that automation-era management requires.

The encouraging part is that these are capabilities anyone can start building now, in any role, and they compound. Getting a head start on governance and judgment, while others are still competing on process knowledge that AI is absorbing, is one of the clearest ways to make yourself more valuable as management transforms.

What Does This Mean for the Future of Work?

The broader lesson is that AI is not eliminating management, it is raising the bar for what management means.

The value is moving up the skill ladder. As AI absorbs the routing and control functions, the human contribution concentrates in judgment, governance and design, which are harder and scarcer. That is the same pattern appearing across many roles, where AI takes the routine layer and the durable human value shifts to what sits above it.

Titles matter less than capabilities. The experts describe a world where the managers who remain are defined by understanding the logic behind systems, not by their position in a hierarchy. In an automated organisation, what you can actually do, govern, decide, design, matters more than the box you occupy on an org chart.

Governance is the emerging scarce skill, and like avoiding the learning debt AI can hide, it rewards those who build depth deliberately. The struggling organisations are those that automated without building governance, which signals rising demand for people who can provide exactly that. For anyone planning a career, the ability to govern intelligent systems responsibly is a bet on where value is heading.

What if You Are Already a Manager Watching This Happen?

Many people reading this already hold a management role and feel the ground shifting, so it is worth addressing that position directly.

Audit where your value currently sits. Be honest about how much of your role is information routing, approvals and reporting, versus judgment, governance and design. The first category is the shrinking one, and knowing your current mix tells you how much you need to shift, without waiting for a reorganisation to force the question.

Deliberately move up the value ladder. If much of your value is currently in control and routing, start converting it toward the durable skills, understanding the logic behind your processes, governing the automated systems entering your team, and owning the judgment calls. That transition is entirely doable, and doing it proactively is far better than having it done to you.

Redesign your own function before someone else does. The valuable manager can redesign the workflow when conditions change. Applying that to your own role, actively rethinking how your team works as automation arrives, rather than defending the old structure, is how you stay ahead of the shift and demonstrate exactly the capability the future rewards.

Lead your team through the same transition. The people you manage face the same bifurcation, and helping them build judgment and governance skills, rather than clinging to process knowledge AI is absorbing, both serves them and marks you as the kind of forward-looking manager organisations most need. Guiding that transition is itself a demonstration of the higher-order management the experts describe.

Treat this as an opportunity to aim precisely. Management is not a dying field, it is a transforming one, and the transformation rewards those who build the harder, judgment-based skills over the routing-based ones. Understanding that distinction, and developing toward the growing side of it, is how you turn an anxious debate about whether managers are obsolete into a clear plan for becoming the kind of manager the future actually needs.

Related Articles

• The Career Development Gap Most Workers Blame on Themselves Is Their Manager
• The 7 Warning Signs That Make Good Employees Start Job Hunting
• What Gets People Promoted in 2026
• Why Accidental Managers Sink Teams
• Managing AI Agents Is Becoming Its Own Job
• The AI Skills Employers Actually Want
• A 90 Day Plan to Build AI Skills
• AI Jobs and the Future of Work in 2026

People Also Asked

Q: Is AI making managers obsolete?

A: Only one kind. The expert view is that AI is making obsolete the manager whose entire value was controlling information flow, because AI now handles that routing function. The manager who governs automated systems and owns judgment calls is becoming more valuable, not less.

Q: Do flatter organizations need fewer management skills?

A: No, they need different and harder ones. Experts found that removing approval layers raised the skill floor rather than lowering it, because the remaining managers must interpret systems, intervene at the right thresholds, and redesign workflows. Flatter does not mean simpler.

Q: Should I still aim for a management career?

A: Yes, if you build the right skills. Avoid basing your value on information control and approvals, the part becoming obsolete, and develop judgment and governance instead, deciding what needs human input, owning escalation thresholds, and understanding the logic behind processes. That is the growing, durable side of management.

Ready to level up? Metaintro helps you see which leadership and AI-era skills the market actually rewards, so you can build toward the management roles that are growing. Create your free profile and plan your path.