The World Series just ended and it was a heart breaker for the Toronto Blue Jays, who were 2 outs away from winning their first in 32 years.

If you’ve been with me for a while, you’ll know that I draw a lot of analogies to baseball…and with the official end of the 2025 Season, I thought I’d get one more in before Winter.

One of my favorite Yankee traditions is Old-Timers Day, where retired players come back and play a game against each other. Throughout the game, former players and managers will visit the broadcast booths.

This past summer Joe Torre, Yankees manager who oversaw the 1996, 1998, 1999, and 2000 World Series wins, said something that struck me as very interesting: that he would manage by “feel.”

While analytics tells you to do one thing or make a certain move, that doesn’t really compare to looking a ballplayer in the eye and seeing what they’ve got left — seeing what they can do.

That, naturally, got me thinking a lot about our relationship with AI.

Bad Things Happen When We Rely on Analytics Too Much

The Yankees just haven’t been the same team in the Analytics era. Many of their losses — and I’d say the reason they got embarrassed in the post season — were more about looking at analytics, matchups, numbers, and not much about feel.

They’re so worried about pitch counts, overworking guys, making the perfect move according to the stats.

But they’re missing that gut-level contact with the game — that sense of knowing what’s really needed in the moment.

It’s become more about avoiding mistakes than making bold, intuitive choices.

It seems like they’re outsourcing some hard decisions to a number-crunching computer.

The Same Thing can be Said About AI

Spend any time anywhere online and you’ll see the same thing with AI.

As we use it more for our work, we need to consider what we’re losing. Sure, AI can give us the statistically likely answer, but it can’t tell us what our gut tells us. It can’t tell us what our heart tells us.

It can’t give us a truly human response.

When ChatGPT-5 came out, and I saw people complaining that the “voice” they trained it to have completely changed. Of course it did.

And AI is not a person, even though we’ve personified it. Its voice isn’t really a voice. Its gut isn’t a real gut.

It is software manipulated by people.

Don’t Lose Your Voice to AI

When you outsource your voice to AI, you risk losing it. You risk it changing because it’s easier than actually doing the work.

My friend recently told me, “There are people who don’t breathe unless AI tells them to.”

It’s funny, but it’s increasingly true. You get a good result and then you feel you can trust it, and stop thinking about what it’s doing for you.

So as you think about how you’re going to use AI in the future, I want you to think about what Joe Torre said. We need to be mindful about why we, as small business owners, use AI: it’s to help us do better.

Yes, analytics are helpful, but it’s no replacement for heart, gut feeling, or knowing in the moment the move you need to make.

That is something that will always be uniquely human.