Teaching Kids to Think First and Ask AI Second
AI will answer any question your child asks, but that doesn't mean it will answer well. The quality of what they get back depends entirely on what they put in. And that's exactly where our teaching opportunity lives.
In my last post, AI Won't Teach Kids to Think — That's Still Our Job, I wrote about why we can't wait for schools or tech companies to teach kids how to use AI wisely. Thinking still starts with conversations about AI that happen while driving, around dinner tables, during homework sessions, etc.
My next set of blog posts are going to explore what that looks like in practice for us all to think first and ask AI second. Each AI exploration is designed for kids from elementary to high school age and is simple (and fun) enough for families to try together. Remember, AI isn't something to fear or blindly trust, it's something to learn with. And the more we explore it side by side with our kids, the better we all become at asking thoughtful questions, testing ideas, and thinking for ourselves.
Here's where we're headed with five explorations:
Asking Smart Questions – Curiosity drives learning—and good prompts lead to good thinking.
Brainstorming as a Spark That Grows – Great ideas come from layering and refining questions.
Fact-Checking and Curiosity (with Lateral Reading) – Don't stop at the first answer; "read sideways" to confirm what's true.
Debating with AI – AI offers perspectives; kids still decide what they believe.
Planning & Organizing with AI – AI strengthens organization, but kids remain the leaders.
Now, let's start the first exploration, where all good learning starts, with questions.
Exploration 1: Asking Smart Questions
When kids first use AI, they often type a quick question - "What's photosynthesis?" or "Who invented the airplane?"- and take the first answer that appears. But AI isn't a search engine; it's a conversational partner that responds to how you ask. The clearer and more thoughtful your question, the better the response.
That's the heart of what's now called prompting. In fact, it's become such an essential part of working with AI that entire careers now focus on prompt engineering, basically how to shape questions that guide AI effectively.
A prompt is what you tell AI to do — a question, direction, or even a role you want it to take. For example:
"Explain gravity like I'm 10 years old."
"Help me brainstorm ideas for a story about time travel."
"Act as a science teacher and quiz me on the solar system."
Each of these prompts will produce a different kind of answer, tone, and depth. In other words, the prompt helps you guide AI. Good prompting teaches kids far more than how to use AI effectively, it also teaches them how to think.
When kids take a moment to plan what they're asking and why, they're learning to express themselves clearly, identify what they need, and make decisions about what information is relevant. They're also learning that if an answer feels off, vague, or too general, they can try again — and that asking better questions almost always leads to better thinking.
How to Write Better Prompts
Prompting isn't something to teach once and move on from - it's something to practice, like any meaningful conversation. Here are some tips about how to help kids develop and practice writing good prompts.
1. Keep it simple, but clear
AI does best with prompts that are short, focused, and specific. If you ask it to do too much at once, for example, "Explain the water cycle, write a poem about it, and make a quiz", it will try to do everything but won't do any of it well.
Encourage kids to break big tasks into smaller, connected prompts. Sometimes it takes a short series of prompts to get to a response that really fits what you need.
2. Be specific
Vague questions lead to vague answers. Instead of "Tell me about ecosystems," try "Explain how a desert ecosystem survives with little rain." The more context you give, the more useful the response.
3. Add the "why"
Curiosity deepens learning. "List three causes of the American Revolution" becomes "List three causes of the American Revolution and explain why each one mattered to colonists."
4. Set the role
Ask AI to "act as" someone — a teacher, coach, scientist, or friend. This not only changes the voice and depth of the answer but also helps kids see that perspective matters in how information is presented.
5. Revise and reflect
Encourage kids to ask AI for a second draft: "Make it simpler." "Add an example I could test." "Explain it like you're talking to a fifth grader." These small tweaks model the kind of keep refining thinking great learners use every day.
Try This: The Prompt Remix
Start with yourself first. Before trying this with your child, pick a topic you're genuinely curious about—how sourdough starter works, why certain songs get stuck in your head, or what makes a good golf swing. Ask AI a basic question, then revise it three times using the tips above. Notice what changes. You'll be amazed at how much better your third attempt is than your first—and you'll have lived examples to share with your kids.
Then, pick a topic your child is curious about, such as volcanoes, constellations, favorite animals, or even why people dream. Start by typing a basic question into AI and read the answer together. Then, challenge them to rewrite that same question three different ways by adding details, examples, or tone changes each time.
Compare the responses and talk about what changed:
Which question gave the clearest or most interesting answer?
What words seemed to make the biggest difference?
How could you refine the prompt one more time?
This small exercise helps kids see that prompting isn't a trick; instead, it's thinking in action.. It's about staying curious, asking clearly, and being willing to keep trying until the answer makes sense. Periodically check in with your kids about the questions they are asking AI and how they are changing prompts to improve the quality of AI responses.
Why This Matters
Every time your child stops to revise a question, they're practicing something AI will never do for them which is deciding what matters.
When your kid asks AI to "explain photosynthesis" and gets a wall of text, then revises it to "explain photosynthesis like I'm teaching it to my little sister", that little shift isn't just about getting a better answer. It's about that child realizing they need to understand the concept clearly enough to teach it. That is metacognition or more simply said, thinking about how you think.
The goal isn't perfect prompts. The goal is for kids to pause before they ask, think about what they really need, and know they can always ask again better. These habits of clarity, revision, and curiosity transfer to every conversation they'll ever have, every problem they'll ever solve, and every decision they'll ever make.
AI isn't teaching our kids to think. But learning to work with AI can teach them to think better if we're there to guide the process.
Next Blog Post
We'll shift from asking better questions to growing ideas. In the next exploration, we'll look at how to use AI as a creative partner that helps kids expand, reshape, and refine their thinking without taking over the process. We'll explore how to use AI as a creativity partner that helps to layer and refine ideas instead of letting it create all the ideas on its own.
If you'd like to follow along with all five explorations, make sure to subscribe below.
A Note on Process
For this blog post, I used ChatGPT to help me synthesize several blog post ideas into a small series of explorations. Then I revised the post using feedback and suggestions from Claude.
