AI and Management: What Changes, What Doesn't

Most managers are drowning in the wrong work.

Not the hard work. The *heavy* work. Summarizing a one-on-one. Writing up a performance review that says what you already know. Digging through your notes to remember what you told someone three weeks ago. Preparing for a calibration session that should take an hour but takes four because you're assembling evidence from scratch.

That's the work AI is actually good at. And if you're not using it for that, you're leaving time on the table that you can never get back.

But here's where a lot of managers go sideways: they either reject AI entirely because it feels impersonal, or they lean on it so hard that they stop thinking. Both are mistakes. The first one wastes your time. The second one quietly erodes your judgment.

Let me be specific about what the line actually looks like.

What AI Can Do Well

AI is excellent at anything that involves organizing, summarizing, or drafting from information you already have.

In practice, that means: transcribing your meetings and pulling out action items, surfacing patterns across performance notes you've written over months, drafting a first version of a review based on your documented observations, helping you prepare talking points for a difficult conversation, and keeping a running record of what you committed to so nothing slips.

These are real things. Not hype, not future-state speculation, but available now and working.

Here's a concrete example. Say you've had twelve one-on-ones with someone over the past quarter. You've taken notes, but they're scattered. When it's time to write their review, you're essentially reconstructing a narrative from fragments. AI can pull those notes together, identify themes, and give you a coherent first draft in minutes. You still decide what matters. You still make the judgment calls. But you're not starting from a blank page, and you're not missing the pattern that was sitting there all along.

That's not a small thing. That's hours back in your week, every week.

What AI Cannot Do

Here's where I'll be blunt: AI cannot build a relationship, and relationships are the job.

I've managed hundreds of people across my career, and the moments that actually changed someone's trajectory were never administrative. They were human. The conversation I had with a guy who was about to quit over something his skip-level said, where I had to read past what he was telling me to get to what was actually wrong. The time I pushed back hard on a strong performer because she was sandbagging, and I had to know her well enough to know she could take it. The manager who was struggling with her team and needed me to sit across from her for an hour and just listen before I said a single word of advice.

No AI does that. No AI knows when to push and when to pull back. No AI reads the energy in a room when someone gives a technically correct answer but something feels off. No AI earns trust over time through consistency, through showing up, through doing what you said you'd do.

The human part of management isn't just important. It's the whole game. Everything else is overhead.

The Time Math

Here's what changes when AI handles the overhead.

The average manager I've worked with spends somewhere between 30 and 50 percent of their time on documentation, administrative prep, and information retrieval. Write that down: up to half your working hours. If AI cuts that by even 60 percent, you've suddenly got 15 to 30 percent of your week back.

What do you do with that time?

The managers who thrive aren't the ones who resist AI or the ones who outsource their judgment to it. They're the ones who use it to buy back time for the work only humans can do: the real conversations, the coaching, the trust-building, the being present when someone needs you to be.

That's the reframe. AI doesn't change what great management looks like. It changes how much time you have to actually do it.

The Danger of Using AI as a Crutch

I want to be direct about something, because I see this pattern already.

If you let AI draft your performance reviews without deeply engaging with the output, you will write reviews that sound plausible but miss the person. Reviews are a form of relationship. The person reading it knows whether you actually saw them this year.

If you let AI summarize your one-on-ones without reviewing the summary critically, you'll lose the details that matter. The specific thing someone said that told you they were struggling. The throwaway comment at the end of a meeting that was actually the most important thing said. AI doesn't know what's significant. You do.

If you stop taking your own notes because AI will transcribe everything, you'll stop processing what you're hearing in real time. Note-taking isn't just documentation. It's how your brain locks in what matters.

The risk isn't that AI replaces your judgment. The risk is that you let it, gradually, without noticing.

What a Good Day Looks Like

Let me paint you two pictures.

Manager A starts Monday with a one-on-one. She walked in without reviewing her notes from last week, because she was planning to "just remember." The meeting is fine but shallow. She spends 45 minutes after it trying to write up what happened. Her afternoon calibration session runs long because she hasn't organized her evidence on three people and keeps losing her thread.

Manager B starts Monday the same way. But before the one-on-one, he took five minutes to review an AI-organized summary of the last three sessions, which flagged that he'd committed to introducing his report to a VP and hadn't done it yet. He walks in prepared. The conversation goes somewhere real. His notes sync automatically. By the afternoon, his calibration prep is mostly done because the documentation work happened continuously, not in a panic. He uses the extra hour to have a conversation he'd been putting off with someone who was quietly disengaging.

Same number of hours. Completely different outputs.

The Actual Skill Now

The new skill for managers isn't learning to use AI. Anyone can learn the tools. The skill is knowing what to give it and what to keep for yourself.

Give it the organizing, the drafting, the summarizing, the pattern-surfacing. Keep the judgment, the relationship, the reading of the room, the hard conversations. Use it to create space for presence, not to substitute for it.

The managers who figure that out early will be operating at a different level than the ones who don't. Not because of the technology. Because they bought back time and spent it on the right things.

That's the whole idea.