Part 4: Grow Leaders and Managers with an AI Team — Directing Is Developing

Series: “Careers & Skills in the AI Era” (5 parts)

1. How Not to Let AI Take Your Career
2. Beginners, Build Experience with AI
3. Make AI Your Student and Escape the Intermediate Trap
4. Grow Leaders and Managers with an AI Team (this article)
5. Those Who Grow AI Win

Introduction

Last time, we talked about intermediates getting past the wall by “teaching AI.”

As you keep “teaching AI,” the next stage comes into view: instead of instructing a single AI, leading multiple AIs as a team. This time, the point is that the experience of leading an AI team is, itself, what grows leaders and managers.

”Instructing an AI Team” Is the Same as “Instructing to Develop People”

First, let’s break down what good managers do day to day.

• Break work down: Split a large task into manageable sizes.
• Make expectations clear: Agree on what, how far, and by when.
• Set priorities: What to tackle first, what to defer.
• Share criteria: Hand over the yardstick for how to choose when in doubt.
• Give feedback: Point out gaps so they can be applied next time.

On teams where new members grow, these five usually turn smoothly. And what you need to get good work out of an AI team is exactly the same.

In other words,

A good manager can also give good instructions to an AI team.

People who can only give vague instructions are equally vague toward subordinates and toward AI. Conversely, people who can clearly verbalize expectations and criteria draw out good results whether the counterpart is human or AI. This convergence is starting to be tested in corporate training (The Times of India), and Mike Taylor (Every) argues that AI management and human management are converging.

An AI Team Becomes a “Practice Ground” for Management

Here’s the interesting part of the AI era. Management used to be impossible to practice until you actually had subordinates. But now, you can gain the experience of leading by treating AI as your team.

• Break a task down and assign it to multiple AIs.
• Give each a role and criteria.
• Review the outputs and course-correct with feedback.
• Integrate the whole into a single result.

This is a miniature of team management itself. And because AI throws back unexpected reactions, misunderstandings, and resistance, it’s quite practical training. (Research on developing leaders with AI is also growing: OUP Academic.) Role-playing 1-on-1s and review meetings with AI as the “subordinate” can be tried anytime, any number of times, at low cost.

“The ability to develop people” and “the ability to lead AI” reinforce each other—train one and the other is trained too. The AI team is an ideal practice ground for growing leaders and managers.

Solo Work and “Collaboration” Are Different Dimensions

Lead an AI team and you notice it fast: working alone and orchestrating several in parallel demand completely different perspectives.

Things you never had to think about solo suddenly become problems the moment it is a team.

• How to prevent deliverables from colliding: if several AIs edit the same file at once, it breaks easily. You need coordination—separate branches, clear ownership boundaries, merge rules.
• How to share information and know-how: unless the assumptions in one head are put in a form the whole team (people and AI) can reference, the same failure gets repeated everywhere.
• How to share heavy machine resources and file systems: who uses builds and shared storage, and when? Without designing for contention, queuing, and cleanup, you jam up fast.

In software terms these are version control, CI, and shared-environment design; in management terms they are role division, information sharing, and resource allocation. They call for a collaboration-specific perspective that is not a mere extension of solo work.

Running an AI team lets you feel this “difficulty of collaboration” firsthand—which is exactly why it doubles as training for team management.

The Leader’s Role Changes

Let’s widen the view one more level, to the whole organization.

Leaders used to often be “the person with the answers”—the most knowledgeable, the most experienced, the final judge. But now that AI provides knowledge and first-pass answers instantly, “having the answers” is no longer what a leader is valued for.

The leader of the future shifts toward “the person who combines AI and humans to maximize results.” (OUP Academic) When the role changes, so do the items to train.

”The Ability to Teach” Becomes an Organizational Capability

Recent leadership research tends to define an excellent leader not as “someone who grows only themselves” but as “someone who raises the whole team’s capacity to learn” (Academy of Management Learning & Education).

Add AI, and a new loop of knowledge transfer is born.

The leader teaches the work to AI
        ↓
AI supports the members
        ↓
The members grow
        ↓
New insight accumulates again as explicit knowledge
        ↓ (loop)

Tacit knowledge used to stay locked inside one person’s head. When teaching AI turns it into explicit knowledge, that knowledge spreads across the whole team via AI and settles into the organization. “The ability to teach” is promoted from a personal skill to organizational infrastructure.

• Traditional manager development: the ability to “develop people.”
• AI-era leader development: the ability to “teach both people and AI.”

This is not just one part of an “AI skill.” It is a core leadership capability that trains structured thinking, design ability, teaching ability, and leadership all at once. Indeed, 2026 leadership research increasingly includes AI literacy among a leader’s core competencies (PubMed).

Summary

• The template of good management (break down, expectations, priorities, criteria, feedback) overlaps exactly with instructing an AI team.
• So the AI team becomes a practical practice ground for growing leaders and managers.
• Solo work and collaboration are different dimensions: avoiding deliverable collisions, sharing information, and sharing heavy resources/file systems all demand a collaboration-specific view.
• The leader’s role shifts from “the person with answers” to “the person who combines AI and humans for results.”
• “The ability to teach” is promoted from personal skill to organizational capability, creating a knowledge-transfer loop.

Next time is the finale. As a synthesis, we dig into the core skill common to beginner, intermediate, and leader: turning experience into a form people and AI can act on.

👉 Next: Those Who Grow AI Win