Raise a Thought Partner, Not a Rubber Stamp

My oldest daughter is a resort management major. Between a part-time job at a private club and running her sorority's event calendar, she's planning something almost every week — mixers, formals, member events, whatever's next on the schedule.

For a while, she'd come to me the same way every time: “I have this event coming up and I'm not really sure what we should do.” Then she'd go ask AI the exact same open-ended question and take whatever it handed back.

I told her to stop asking and start directing. Before she asks for a single idea, she hands over:

•       Audience — who's actually going to be there.

•       Timing — season, time of year, day of the week.

•       Track record — what's worked before, what's flopped.

•       Constraints — budget, venue, anything non-negotiable.

Then she asks for recommendations — and once she has them, keeps going. Follow-up questions. More context. Pushing on the parts that don't quite fit her crowd.

That makes AI her planning assistant, not her ghostwriter — the same way I'd lean on a sharp coworker to pressure-test an idea before I committed to it. That's a real skill. And it's going to separate her from another candidate with the same degree and the same GPA, long after graduation.

A few issues back I wrote about the difference between treating AI like a search bar and treating it like a coworker you actually brief. This is the next layer of that same idea, and it matters more the younger the user is.

The habit worth building — for my daughter, for me, for anyone — isn't asking AI a question and moving on. It's acting like a thought leader in the conversation. You set the direction. You give the context only you have. And when the answer comes back, you don't take it as finished. You challenge it. You ask why. You push back on the parts that don't sound right, the way you'd push back on a report from someone on your team before you'd put your name on it.

I learned that lesson on a flight line, working as an F-15 crew chief, long before I ever learned it at a keyboard. Nobody signed a jet off as good to go because someone said it probably was. You verified it yourself, every time, because “probably fine” isn't a standard you fly on. Trust but verify isn't a slogan I picked up from a business book. It's how I was trained to think, and it turns out it's exactly the muscle AI rewards if you actually use it.

That posture drives a better output. It also turns a five-minute AI exchange into an actual conversation, the kind where you're thinking harder about the problem by the end of it, not less. That's true whether the person on the other end of the chat is 55 or 20.

I don't think I'm the only parent noticing this. A few studies landed in just the last few months, and together they say something more useful than either headline on its own.

RAND's American Youth Panel found that as of last December, 67% of students agreed that heavier AI use on schoolwork is actually harming their critical thinking. Kids are clearly using AI already. What that number actually measures is how.

Child Trends found something that should reassure any parent wrestling with this alone: 57.6% want responsible AI education built into how their kids' schools already teach, not treated as a bolt-on elective. Parents aren't asking schools to ban it. They're asking someone to teach the how.

And a recent Guardian survey found 40% of K-12 teachers now say their students use AI in the classroom at least once a week. That number isn't heading down. Whatever your school district's policy says on paper, the tool is already in your kid's hands.

Put those three findings together and one thing is clear. AI is already in front of kids, whether their school welcomes it or looks the other way. What's missing is someone teaching them the difference between asking and directing. That's a gap a parent can close at the kitchen table, no permission slip required.

Next time your kid — or you — is about to ask AI something open-ended, try building the ask like this instead:

“I'm working on [task]. Here's who it's for: [audience]. Here's what's worked before: [past context]. Here's what's different this time: [constraints or budget]. Give me a few directions, and I'll tell you which one to build out.”

Fill in your own blanks. Here's roughly what my daughter's version looks like when she's planning a club mixer:

“I'm planning a fall mixer for about 60 club members, mostly 20s and 30s. Past events that worked well had a theme and a signature drink; the ones that flopped felt generic. Budget is tight this time. Give me three theme directions with a rough setup for each, and I'll pick one to flesh out.”

Then don't stop at the first answer. Ask why it suggested what it did. Tell it what's missing. Make it earn the final version instead of accepting the first one.

Parenting through this stretch is hard enough without adding a technology fight to the list. But pretending AI isn't already part of your kid's life doesn't protect them — it just means someone else is teaching them how to use it, or no one is. The parents who lean in now, even imperfectly, are giving their kids a real head start into a workforce that's going to expect this fluency as a baseline, not a bonus. Skipping it doesn't keep a kid safe. It leaves them behind while everyone else gets fluent.

Hit reply and tell me how your family is handling this one. I read every response.

See you next Monday.

— Eric

More of this on LinkedIn: linkedin.com/in/eric-skelton-mba-3079a47