Helping Kids Navigate AI? Think of It Like Streaming Services
Every night, it’s the same scenario: Kids are finally in bed (well, they may come out 1-2 more times, but they are mostly in bed), husband and I are ready to unwind and we just want to relax and watch something on TV.
It should be so simple.
But. It. Never. Is.
Without fail, we go through streaming service after streaming service.
Netflix. Hulu. Max. Apple TV. Amazon Prime. Peacock.
We know we saw something somewhere we wanted to watch, but can’t find it. Most of the time we just settle on whatever and then end up scrolling on our phones. Other times we give up. Sometimes, if lucky, we find something and enjoy it.
Does this sound familiar? You’re paying for six services, using two and still can’t find anything to watch.
Guess what? This is the same feeling many parents feel when trying to pick an AI tool to use, especially one to use with their kids.
ChatGPT. Claude. Gemini. NotebookLM. DALL-E. Midjourney. Copilot. Perplexity.
So. Many. Chatbots.
We are at a critical juncture where we realize AI is here to stay. And, for that reason, we need to start teaching our kids to use it with intention. Otherwise, it becomes a shortcut students use so they don’t have to write that paper or read that book or figure out that math equation.
But where do we start?
Here’s the trick. Despite your initial inclination, do not start with the tool. Just pick one generalist tool and go from there.
After using AI with my 5- and 7-year-old for the last six months, I’ve learned:
The tool doesn't matter nearly as much as what you're teaching in that moment.
Let me give a couple of examples.
The next great superhero
Most recently, I used AI with my kids to create a superhero who didn’t yet exist, but they thought should.
So, I had them both describe to me their character by answering questions like:
Is your superhero a villain or good guy?
What is their super power? How did they get it?
Who are they fighting and how do they win?
What happens when they’ve captured the bad guy?
Do they have a #1 enemy?
I typed out everything they told me. They answered questions I didn’t even ask. Then, at the end of the prompt, I asked AI (in this case ChatGPT) to generate an image of what this superhero looked like in action.
Image 1: Too generic, kids made modifications and additions
Image 2: Closer, but still had some flaws
Image 3: Okay, we can work with this
Even after it was printed, they drew on it and kept making changes.
During this they learned: It was all about their imagination. They were in charge, not the other way around.
And the tool didn’t matter. I could have used Gemini. I could have used MidJourney. It didn’t make a difference, the lesson would be the same with any of those.
The recipe that shouldn’t have been
Another time, my kids picked four ingredients and asked our friend Claude (by Anthropic) to create a recipe.
The ingredients? Strawberries, cheese, chocolate sauce, elbow macaroni.
AI gave us “Strawberry Chocolate Cheese Bites.” And we made them. My kids spit them out.
Again, it didn’t matter what chatbot we used. They learned: AI will try to make anything work, so it’s our responsibility to figure out what goes together or not. AI doesn’t “think.” it predicts.
As an added bonus, they realized that just because they like each item individually, doesn’t mean they should be mixed together.
So, do you get where I’m going?
The question isn’t, “what tool should I use?”
The question is “what do I want my kid to take away from this?”
The four main principles (in my very humble opinion) when it comes to kids and AI
Over the next few weeks, I’m breaking down the core principles I focus on with my kids. Yes, I will share some general tool recommendations along the way. Because some tools just don’t work for some use cases.
AI makes mistakes. And we learn from questioning outputs. AI will try to people please at all costs. And it will make stuff up!
Your Imagination and ideas need to come first. Kids lead, AI follows. To get what you want out of it, you have to give it lots of information (from your own brain) and direct it as it generates.
AI is a tool. It is trained on data and processes information really fast. It’s not the artist. It’s a compiler of art. It is like a really smart librarian. It knows where are the good books are, but didn’t write them.
The AI output is not the finish line. Bring it into the real world. No matter how many iterations you go through in the tool, don’t let it end there. AI should be in the middle of the cycle, not at the beginning or the end.
This isn’t a tool comparison. It’s a framework that will work years from now, no matter how many AI tools we have. It’s not about mastering the tech, it’s about raising kids who can think critically in an AI-present world. It’s about defining what should always be done by humans, and just amplified by tools.
Let’s get back to streaming services
Just like with streaming services, there are many options. And you’ll likely try man. That’s normal. Then you’ll settle into your regulars:
Chat GPT might be your Netflix: The everyday go-to
Claude, your Apple TV+: When you want something a little more thoughtful
NotebookLM could be your Peacock: You forgot you have it until you need that ONE specific thing, like a binge watch of Bravo Housewives.
And, yes, you'll keep a few others around for those random moments when only that specific tool will do. That's fine too. The point is: you don't need to master them all before you start. Pick one. Begin there. Let your actual needs (not FOMO) guide which others you add. The principle you're teaching matters more than the platform you're using.
If you’re a parent figuring out where to start with AI and your kids, follow the series.
If you work with AI in your professional life, I bet you’ll relate too.