The AI-Native Leader, Part 1: AI Doesn't Need Fewer Managers. It Needs System Thinkers.
Part 1 of a 3-part series, "The AI-Native Leader." AI isn't replacing management — it's revealing what management should have been all along.
These two questions have been coming up more and more in my coaching lately:
From strong IC leaders: “I was ready to move into management — but now I’m not sure. Should I even bother?”
From experienced managers: “I don’t think we’re going to need as many managers in the future. Should I go back to IC?”
Different situations, same anxiety. And here’s the thing — the second question isn’t wrong. It’s practically consensus at this point that we won’t need as many managers. But both questions are still pointing at the wrong problem.
The real question isn’t “manager or IC.” It’s: were you doing system design, or were you doing paper logistics?
What good management has always been
Before we talk about AI, let’s talk about what good management actually looks like — because not everyone frames it this way, and I certainly didn’t when I started.
A good manager is a system designer. That sounds clean in a sentence, but in practice it’s messy — and most of us learn it the hard way. I know I did.
Think about the difference between coordinating and designing:
Your team keeps getting blocked by another team. The coordination instinct is to escalate, chase people, add meetings. The design question is: why does this keep happening? Have a real conversation with the other side, understand their constraints, and redesign how the teams interact.
Everything depends on you or a few seniors to decide. The coordination instinct is to work longer hours. The design question is: can you push decisions to the right level and clarify what actually needs your judgment versus what doesn’t?
The same issues keep repeating — missed deadlines, misalignment, quality drops. The coordination instinct is to jump in and fill the gap yourself. The design question is: what needs to change so the team handles this without you?
These aren’t things bad managers do versus good ones. Every manager makes these mistakes. The shift to system thinking isn’t a talent — it’s something you develop, usually after enough bruises.
I’ll share an example.
I coached a manager who was, by most measures, doing everything right. Strong culture, cared deeply about his people, worked tirelessly to unblock the team. Everyone liked working for him.
Then his team’s charter expanded. More scope, more pressure. Things started breaking.
The team was built around a few high performers covering for the rest. That worked when the system only needed to handle one major effort at a time. But under pressure — multiple priorities, competing demands — the system failed.
His instinct was what it had always been: jump in, support harder, shield the team.
But the real issue wasn’t effort. It was design. The team wasn’t built to handle the new load. Culture alone doesn’t make a team that can scale. Building a balanced, capable team — one that holds up when the environment changes — that’s system design.
What AI actually changed
So what changed with AI? Not the goal. But the system you’re designing now has a new kind of player in it — one that moves fast, never pushes back, and has no stake in the outcome.
Think about what that means for delegation. A huge part of a manager’s week has always been information logistics: translating context between levels, synthesizing status, coordinating handoffs, shepherding decisions. That work is folding. AI summarizes, translates, drafts, pulls data — the mechanical layer compresses.
So the delegation question shifts. It’s no longer just “which human owns what.” It’s: which work goes to AI, which work still needs a human, and where are the handoffs between them? If the coordination was the job — if that’s what filled your days — then the honest answer is that work is folding. Not because you did it poorly. Because AI does it faster.
And the system exposes itself quickly now. If goals are unclear, you get more output but not better. If ownership is fuzzy, confusion spreads. If quality standards are weak, bad work scales. AI amplifies whatever system it’s plugged into.
AI doesn’t make management obsolete. It makes coordination-as-management obsolete.
Get your hands dirty — but for the right reason
Here’s where it gets practical: you need to understand what AI actually does in the context of your team’s real work. Which tasks compress. Where judgment still matters. Where the handoffs break.
You can’t figure that out from a distance. And the two reactions I see most often both miss: diving back into execution yourself, trying to be the most productive IC again. Or staying hands-off and mandating outcomes from above — “Let’s improve productivity by 30% by Q3” — without understanding how the work actually changed.
Get your hands dirty, but not to become a super IC. Start with your own paper logistics — delegate the status synthesis, the context translation, the document prep. Then pair with a team member on a prototype. Use AI end-to-end on a real deliverable. You’re doing this because you’re the system designer, and you need to understand the new material before you can redesign around it.
That’s not new. Good managers have always done this — gotten close enough to understand the real constraints, then stepped back to shape the system. The material changed. The job didn’t.
The reframe
So let me come back to the two questions I opened with.
To the manager considering IC: if the coordination work was what you were drawn to — and it’s folding — the IC path has never been stronger. ICs with AI have more leverage than ever. Going IC isn’t a step backward. It might be the most honest move you can make. Either way, make that decision from desire, not fear.
To the IC considering management: this isn’t about a title. It’s about whether you want to design the system or work within it. If you get energy from figuring out how teams should be built, how decisions should flow, how an organization adapts — the AI era needs that thinking desperately. The people who can do this well have always been rare — and the challenge just got harder.
The question was never “manager or IC.” It’s: do you want to work on the system, or work within it?
Next in this series: The Human + AI Harness Model — how to design the layered system your team and org operate in. Subscribe so you don’t miss it.
About Amy Wu
I’m an executive and life coach who works with leaders navigating inflection points — including the one AI is creating right now. If this series resonated, or if you’re working through these questions with your own team, I’d love to hear from you.