StemPro

All articles

July 16, 2026

5 min read

· Learning · AI · Method

AI With You, Not For You: Making Sure AI Doesn’t Do Your Teen’s Thinking

The biggest problem with AI in education isn’t that it’s bad. It’s that it’s too helpful. It will happily do the hard part for your teen — and the hard part is exactly where learning lives.

Where learning actually happens

Think back to something you truly understand. You almost certainly earned it through some struggle: a problem that didn’t work the first time, a bug you chased for an hour, a concept that only clicked after you got it wrong. Researchers call this productive struggle, and it’s not a bug in learning — it’s the mechanism. Skip it and the understanding never forms.

AI can erase that struggle in one click. Or, used deliberately, it can help a student push through it and understand more deeply than they could alone. Those are wildly different outcomes from the same tool.

The principle: with you, not for you

The rule we teach is simple to say and demanding to live: use AI to think with, never to think for you. In practice that looks like four moves:

  • Frame it first. Before touching AI, understand the problem and how it could fail. This is where real thinking starts.

  • Draft, then interrogate. Let AI produce a first pass — then make it justify every choice. Why this approach? What breaks it?

  • Triangulate. Check the output against reality — data, a second source, a mentor. Never trust one answer.

  • Close the AI and rebuild. Reproduce the core yourself, from scratch. This is the tell.

If you can’t rebuild it without the AI, you haven’t learned it.

How to check it at home

You don’t need to understand the technology to test whether learning happened. Just ask your teen to explain what they built — and then to rebuild the core of it without the AI open. If they can walk you through why each decision was made and reproduce the important part on their own, the learning is real. If they can only run the thing, it isn’t theirs yet.

That’s the whole philosophy behind how we mentor: AI is in the loop, never in the lead. The student does the thinking; AI accelerates it; and at the end, the skill belongs to the kid — not the tool.


See how we mentor — the method