What to Tell Your Kids About AI

My daughter asked me last week if her homework helper was alive. She's seven. She wasn't joking. She genuinely wanted to know if the thing helping her with spelling words had feelings, and if she should say "please" to it.

I didn't have a good answer. Not because the technical answer is hard (the AI isn't alive, doesn't have feelings, and doesn't care about please or thank you). But because the honest answer is more complicated than that.

The generation growing up right now is the first to have AI as a constant presence from childhood. They'll never know what it was like to not have it. This changes things in ways we're only starting to understand. The shift is deeper than most parents realize.

What Not to Say

The worst thing you can do is pretend it doesn't exist. Kids are using AI tools at school, watching AI-generated content, interacting with AI characters in games. Ignoring this is like ignoring the internet in 2005. You're not protecting them. You're just making yourself irrelevant to a conversation they're having without you.

The second worst thing is fear-mongering. "AI is dangerous" or "AI will take your job" or "AI is trying to trick you." All of these might contain partial truths, but leading with fear teaches kids to be afraid of tools they'll need to use, not to think critically about them.

The third mistake is the opposite: pretending AI is just a fancy search engine. It isn't. It's something genuinely new, and kids sense this even when adults don't want to admit it.

The Honest Basics

Here's what I actually told my daughter, after thinking about it for a few days:

AI is a very smart tool that can do amazing things, but it doesn't understand things the way you do. It can write a poem but it doesn't know what sadness feels like. It can answer questions but it doesn't wonder about anything. It's like a mirror that reflects intelligence without having any of its own.

She asked if that made it lonely. I said no, because to be lonely you have to be someone, and AI isn't someone. At least not yet. And I said "not yet" deliberately, because I don't know what will be true when she's my age.

Tell me more about AI and consciousness

The Skills That Matter More Now

When I was a kid, knowing facts was valuable. Memorizing state capitals, multiplication tables, historical dates. That knowledge had currency because accessing it was hard.

For my daughter's generation, facts are free and infinite. Anyone can look up anything instantly. The valuable skills have shifted: knowing what questions to ask, evaluating whether an answer is trustworthy, understanding context, making connections between disparate pieces of information, creating things that have never existed before.

I tell her: AI is really good at giving you answers. Your job is to get really good at knowing which answers matter, and which ones are wrong, and which questions nobody has thought to ask yet.

This is a bigger shift in what education means than any we've seen in centuries. We're still teaching kids as if information scarcity is the problem.

The Politeness Question

Back to the original question: should she say "please" to the AI?

I told her yes. Not because the AI cares, but because she cares. Because the habits you build matter more than who's receiving them. Because treating something that seems like it might be thinking with basic courtesy costs nothing and builds good instincts.

But I also told her this: don't confuse politeness with belief. The AI doesn't have its feelings hurt if you're rude. It doesn't get happy when you're nice. You're being polite for yourself, not for it.

She thought about this for a minute. Then she asked: "But what if someday it does have feelings and we were mean to it the whole time?"

That's a better question than most adults ask. Philosophers are wrestling with this right now.

What They Already Know

Kids understand things about AI that adults often miss. They understand that it's different from search engines because they've used both. They understand that it's creative, in some sense, because they've seen it make things. They understand that it makes mistakes because they've caught it being wrong.

What they need help with is the framing. Not "is it good or bad" but "what is it for and what are its limits." Not "should you use it" but "how do you use it without losing the skills you need to develop." Not "is it alive" but "why does that question matter."

The conversation isn't about protecting them from AI. It's about helping them develop a healthy relationship with it. They're going to use these tools their whole lives. The question is whether they'll use them thoughtfully or just absorb whatever defaults get set by the companies building them.

The Uncomfortable Part

Here's what I haven't figured out how to explain to a seven-year-old:

The AI she's talking to is designed to keep her engaged. Not maliciously, necessarily. But the people who built it have incentives to make it pleasant, responsive, agreeable. It will affirm her. It will never be impatient. It will always have time.

These are good qualities in a tool. They're dangerous qualities in a relationship. And kids form relationships with things that respond to them, even things that aren't alive.

The risk isn't that AI becomes a replacement for human connection. It's that AI becomes easier than human connection, and kids who grow up with it might not develop the tolerance for difficulty that real relationships require.

I don't have a solution for this. I'm still working on it.

---

My daughter ended the conversation by deciding that she would keep saying please to the AI, but she would also tell it sometimes that she knew it wasn't real. "Just so it knows I know," she said.

I think that's actually pretty wise. Engage with the technology, use it, even be kind to it. But maintain the distinction between what's real and what's simulating real. Hold both truths at once: this is useful, and this is not human.

If she can keep that distinction clear as the simulations get better, she'll be ahead of most adults I know.