What's a Manager's Job Now That AI Can Answer Most Operational Questions?

A manager's job now is the same job it's always been: building trust, developing judgment, and putting people in the right seat to do their best work. The Agentic Leadership Framework calls that capacity human elevation — helping people grow, contribute, and accomplish more together than they could alone. AI turned six-hour tasks into six-minute ones. It didn't change what management is for. The work is moving faster than the people inside it can read — which makes the manager who develops people more essential, not less.

If your team just got dramatically better at the work — and you're lying awake doing math on who still has a seat when it's done — this one's for you. Stay with me. The answer isn't the one you're bracing for.

The coach can't play the game. The coach can lose it.

A manager I coach asked me a question last month he couldn't ask anyone else. One person on his team — competent, trusted, well liked — did work he could now see AI replacing. If he asked her to help automate it with AI, would she really pour herself into designing an agent that would replace her? Or, if he let her go first, would the team lose the only person who knew how the process actually worked? He couldn't bring the question to his boss. He didn't want to bring it to a chatbot. His wife was tired of hearing it. So he brought it to me — not is this normal, but what do I do?

It's the right instinct and the wrong question. He was trying to solve her problem — figure out the move, make the call, play her position for her. That was never the job. Nearly twenty years of coaching managers lands me on the same point again and again: the manager's contribution isn't executing the play. It isn't carrying the ball and scoring the TD. It isn't throwing the pass. It isn't blocking the tackle. It's seeing the shape of the field that the players can't see, and closing the gap between where this specific person is and where the game is going.

The right question instead looks like this: What can I do to make sure she stays useful, productive, and engaged in this company? What roles can be created for her if she automates her current job? What work is there that we want to be doing yet haven't had the bandwidth for? Can she play and win in those positions?

Yes, AI has changed the field of work. Team members will have to shift positions and tactics. Yet the manager's job — the coach's job — has not changed. It is still to read the field, to develop the strategy, to pivot when needed, to move players around, and, yes, at times in and out of the game. That's not a burden AI added to your role. It's the role — with the volume turned all the way up.

That volume may have you thinking — I'm not sure I know how to do this anymore. Yet it is worth sitting here for a minute and really unpacking why this question is coming up.

The job didn't change. The proof did.

For years, team output was the proof of managerial performance. Your team shipped good work, and good work spoke for you. Now, thanks to AI, most output looks good — polished, fast, repeatable — so the people evaluating your performance are looking somewhere else for proof. What they're looking at has a name. Human elevation is the disciplined commitment to helping people grow, contribute, and accomplish more together than they could alone.

That work was always yours. It's just never been this visible.

Look at what happened at UBS. In 2024 the bank started letting an AI system approve loans on its own. The credit officers didn't vanish. Their job changed shape. They stopped making each individual call and started defining the parameters, stress-testing scenarios, and coaching the system that now makes the calls for them, as Harvard Business Impact documented. Same people. Same trust required. A different game entirely — and someone had to help them see it that way before they could step onto the new field.

Notice what that asks of you. Figuring out what your people do next isn't a memo you write alone. It's brainstorming with them. Innovating with them. Collaborating on what the freed-up hours are actually for. A team can only do that well when two things are already in place: trust, and the skills to think together. If your team is missing either one, everyone's in a little bit of trouble — and there's no tool on your stack that fixes it.

Someone on your team is about to ask you a question you can't outsource.

Your people aren't wrong to feel the ground move. Gallup finds that 18% of U.S. employees think it's likely their job gets eliminated by automation or AI within five years — a number that climbs to 23% once their own organization has actually implemented it. They're doing the math correctly. The question is who's on the sideline while they run it.

You already know who's most exposed. You've probably started running the names in your head, and you hate that you have. That's not a management failure. That's the weight of the actual job, arriving at full size.

Back to the manager I was coaching. Here's where we landed. If she's good — if there's still a position on the field for her — she doesn't disappear when her tasks do. She helps automate the work, and then she manages the agent. Someone still has to. Humans elevate AI. And humans have to elevate humans, because when the day-to-day tasks people built their identity around change this fast, purpose takes the hit before the org chart does.

There's no AI tool for the conversation that comes next — the one where you tell someone her role is genuinely changing, and help her see a next chapter she can't picture yet herself. That conversation runs on the trust you spent years building. Or it runs on nothing, if you didn't. This is the moment that trust either pays out or reveals it was never really there.

And somewhere in the back of your mind, you're running your own name through the list too. That's not paranoia. Gartner predicts that through 2026, 20% of organizations will use AI to flatten their structure, eliminating more than half of current middle-management positions. You're allowed to notice that. Just don't let it be the thing you manage from.

The managers who come through this well won't be the ones who saw their own name on the list first. They'll be the ones who kept coaching anyway — because they understood, before the season started, that the coach who reads the field is the one the players and the front office both need more of, not less.

That was always the job. You've just never had to do it with the field moving this fast.

The managers who come through this well won't be the ones with the cleanest AI workflow. They'll be the ones whose people trusted them before the ground started moving. That's the capacity I unpack across what AI can't replace in leadership — and it's exactly what I work through, one to one, with people who are good at their jobs and done pretending this part is easy.

Private AI tutoring and coaching for experienced professionals navigating what comes next — not a course, not a cohort, a real conversation.