How Leaders Use AI (And Where It Gets in the Way)

The question is no longer whether you use AI as a leader. Almost everyone already does. The question is what for.

That is where the real difference shows up. Some things get better when you bring AI in. Other things get hollowed out. And which is which shapes how your people see you. That image feeds straight back into how effective you can be.

Here is how I look at it with the leaders who come to me. AI takes over the technical and analytical work. That gives you time for the part of leadership only you can do. Being present. Asking the right question. Making real contact. How leaders use AI decides whether that room opens up or disappears.

Start With the Planning

Plenty of operational leaders drown in rosters and capacity questions. How many people do you need, and when? How busy will next month be? How much margin do you build in for illness or overrun?

This is exactly the kind of work AI is strong at. Feed it your historical numbers and let it look for patterns. Which weeks always peak? Where are you structurally short? You get a first analysis in minutes that used to cost you an afternoon.

The output is a proposal, not a decision. You know the people behind the numbers. You know one team member just came back from leave and needs to ease in. You add that context yourself. AI calculates, you weigh.

Read What Your Engagement Data Tells You

Most managers look at an engagement survey and see a grade. A 7.2. Slightly higher than last year. On to the next agenda item.

There is far more in there. Drop the anonymized results from the past few years into an AI tool and ask for trends. Where does the score drop consistently? Which themes keep coming back in the open answers? When did the decline start?

Then ask yourself the uncomfortable question. What does this data say about me as a leader? According to Gallup, only around ten percent of working people feel genuinely engaged at work. Your team is part of that figure. The numbers often show months in advance that someone is checking out, well before the resignation letter lands on your desk.

You do not have to fish those signals out of a spreadsheet yourself. That is what this tool is for.

Let an Agent Prepare Your Conversations

Here it gets more interesting than a tool that summarizes something. You can have an AI agent gather the loose signals before every one-on-one. What has happened since last time? Which agreements were still open? Which themes kept coming up in earlier conversations?

The clever part is the autonomy. A good agent pulls from different sources on its own, cross-references your calendar, and hands you a short briefing. You can set it to run every morning without asking. The briefing is simply there before your conversation starts.

You can also connect an agent like this to your email. Ask for an overview of every message you have not replied to yet, and the messages where you are still waiting on a reply. Have them ranked by urgency. Then let the agent add the open points straight to the agenda of your one-on-one with the right person. That way every loose thread comes back at the moment you are speaking to that person anyway.

And notice what you win that time for. You waste no minutes at the start of the conversation. You do not have to search for where you left off. You walk in prepared and give your full attention to the person across from you. The agent does the digging beforehand so you can put your phone away during the conversation itself.

One warning. An agent that digs through your systems touches your people's personal data. So know exactly which tool does that, and whether it happens safely and within the rules. More on that below.

Help Yourself Set Goals

As a leader you have two kinds of conversations about goals with your people.

In the first kind, you set the direction. "I expect ten percent more revenue this year. Projects X and Y are done by the end of Q3." Clear expectations, clear boundaries.

In the second kind, you flip it around. Here your people come with development goals that matter to them. What do they want to learn? Where do they want to be in two years? You have that conversation together, and it comes from them. A classic coaching conversation.

For the first kind, AI can help you. Let it think along on a sharp wording of a goal. Ask for the blind spots in your reasoning. Check whether your expectation is concrete and realistic. In the second kind, your employee holds the wheel. Let them reach for AI themselves to sharpen their ideas. As long as the goal comes from them and they keep the lead, you strengthen their ownership of that development.

Guard the Line on Personal Data

Here it gets serious. The moment you type a team member's name into an AI tool, you enter personal data. Under the GDPR you are responsible for that. Recent research shows that nearly thirty-five percent of what people paste into these tools contains sensitive information. Much of it without the user noticing.

The solution is simple. Anonymize before you enter anything. Talk about "team member A" instead of a name. Say "an employee with a dip in their numbers" instead of details that point to one person. The analysis stays useful. The person stays protected.

Make a habit of this. It belongs to how you treat your people.

The Email You Are Better Off Writing Yourself

And then the part where AI can truly get in the way of your leadership.

Imagine receiving this from your manager, after a hard week:

"Dear team member, I just wanted to take a moment. Not just any email — but a heartfelt moment of appreciation. Your dedication over the past period wasn't merely impressive; it was nothing short of transformative. In a world that's constantly changing, you are the one who makes the difference. Your contribution isn't just work. It's a journey. Together we build synergy and empower one another to elevate our shared objectives to the next level. Keep shining! Warmest possible regards."

You feel right away what is wrong. Nobody talks like that. It is smooth, correct, and completely empty. No human wrote this.

Research from the University of Florida among more than a thousand professionals confirms the feeling. When people notice that a personal message was heavily AI-generated, the share who see their manager as sincere drops from eighty-three percent to somewhere between forty and fifty-two percent. That same email costs you trust, integrity, and authority.

Reading an AI email feels impersonal. A bit like putting your phone face-up on the table during a one-on-one with your direct report. Both pull the connection away at the moment it matters most.

So write your personal messages yourself. Or if you use AI for a first draft, read it back critically. Does this sound like you? Would you say it out loud this way? If not, rewrite it until it does. A grammar check is fine. Having a machine put your appreciation into words does not belong there.

---

The pattern under all these examples is the same. Use AI for the calculation, the analysis, the first draft. Keep the human work with yourself. The more time you win on the technical part, the more room you have for the part that makes you a leader.

That is also exactly what makes your role future-proof. As AI takes over more technical tasks, the value of a leader moves toward the human. Being present. Having difficult conversations. Asking the right question at the right moment. None of that can be automated.

If you want to get sharper at this, that is exactly what leadership guidance focuses on. For leaders at director level, executive coaching is a logical deepening. And for organizations that want to train their managers in this structurally, a leadership program is a next step.

Running into this yourself and want to talk it through? Plan a free introduction here. We look together at where you stand and what helps you forward.