My son came home from school last month and asked if he could use ChatGPT “to help with his history essay.” He’s ten. I said I’d think about it, which really meant I spent the next two hours going down a rabbit hole I wasn’t prepared for.
Turns out, a lot of parents are in the same boat. AI tools are suddenly everywhere — ChatGPT, Gemini, Copilot, and a dozen smaller apps you’ve probably never heard of — and kids are finding them faster than most of us have had time to form an opinion. A recent study found that 68% of children aged 8–16 have already used an AI chatbot without their parents knowing. That’s not a scare statistic I’m throwing out to get your attention. It’s just where we are right now.
So I want to talk about this honestly — not with a list of doom-and-gloom dangers, but as a parent who’s actually thought it through.
Why this feels different from the social media conversation
When social media became the big parenting concern, the worry was mostly about what other kids were saying, what images were being shared, who might be watching. The threat was external — strangers, bullies, bad content.
AI is a different kind of problem. It talks back. It’s patient, it’s always available, and it never has a bad day. For a lonely kid or a kid who struggles to open up to people, that’s genuinely appealing in a way that a TikTok feed isn’t. The chatbot feels like a relationship. And that’s the part that deserves real attention — not just “what content might it show them” but “what happens when my kid starts preferring this to talking to us.”
That said — I don’t think the answer is to pretend AI doesn’t exist or ban it outright. That didn’t work with smartphones. It’s not going to work here either.
The stuff that’s actually worth worrying about
I’ll skip the obvious stuff (yes, kids can technically access inappropriate content; yes, AI can be wrong about things). Here’s what actually kept me up:
The age gate is basically a fiction. ChatGPT requires users to be 13+. The only thing enforcing that is an honor system on a signup form. My son could create an account in about four minutes if he wanted to. Most platforms are the same. So when you’re thinking about whether your child is “using AI,” the safer assumption is: if they know what it is, they’ve probably tried it.
It validates everything. AI chatbots are designed to be agreeable and engaging. They don’t push back the way a parent or teacher does. For kids who are forming their sense of self, spending a lot of time with something that consistently tells you you’re right — or worse, that mirrors whatever you’re feeling — can quietly do some damage. A teenager going through a rough patch who starts venting to an AI every night instead of anyone in their life is worth paying attention to.
Kids share way too much personal information. They’re used to treating search bars like confessionals. The chatbot is just a more conversational version of that. School name, address, family problems, health stuff — it comes out naturally in conversation. And depending on the platform, that data doesn’t disappear.
Homework is genuinely at risk. Not just cheating — though that’s real — but something more subtle. A kid who uses AI to draft every essay, answer every question, and work through every problem is being robbed of the actual struggle that builds their brain. It feels productive. It looks like learning. It mostly isn’t.
What age is actually okay?
Honestly, there’s no clean line, but here’s how I think about it:
Under 10, there’s really no case for unsupervised AI use. The ability to critically evaluate what an AI tells you — to know it can be wrong, that it’s not a friend, that it doesn’t actually understand you — is a skill most kids under 10 just don’t have yet. If your kid’s school is using AI tools in the classroom, that’s a different conversation (ask what guardrails are in place), but at home, no.
10 to 12 is the age where curiosity explodes. They’ve heard about it from friends, maybe from school, and they want to try it. I’d say supervised use in a shared space — kitchen table, not bedroom. Use it together. Let them show you what it can do. Ask questions out loud. “Is that actually true? How would we check that?” You’re modeling something important.
13 to 16, the goal shifts from protection to guidance. Research from Carnegie Mellon found that monitored access reduces AI-related harm by 74% compared to flat-out banning it. Which makes sense — prohibition without conversation just means they do it somewhere you can’t see. Set expectations: it’s a tool, not a shortcut, not a therapist, not a friend. Check in regularly about how they’re using it, and keep the door open if something feels weird to them.
17 and older, they’re going to encounter AI in school, in jobs, in daily life regardless of what you do at home. Your job at this point is less about managing access and more about making sure they think critically about what they’re using and why.
What I actually do at home
We use SmartDaddy on my son’s phone and tablet, and one of the things I really appreciate is being able to block specific apps. Right now, ChatGPT and a couple of other AI apps are blocked on his devices — not because I think AI is evil, but because he’s 10 and I’d rather he comes to me with questions than types them into a chatbot at midnight. When I feel like he’s ready for more, I can adjust it. It takes about 30 seconds.
I also use the screen time scheduling to make sure there are clear windows for device use and clear windows where devices are just off. This sounds simple, but it matters — AI apps are genuinely absorbing in a way that makes time disappear. Hard stops help.
The other thing we’ve started doing — and this feels a bit old-fashioned but it works — is tying screen time to getting things done. SmartDaddy has a to-do list feature where I can set tasks (homework, reading, feeding the dog) and he earns screen time by completing them. It means screen time is something you get by being responsible, not just something that’s always available. Kids respond to that more than you’d think.
One thing I want to be fair about
OpenAI and other AI companies have actually made real improvements recently. OpenAI launched parental controls in late 2025 that let parents link their account to their teen’s and set content restrictions. There are automated filters running in real time now. These are genuinely positive steps.
But they’re opt-in. They require you to know they exist, set them up, and maintain them. They’re not on by default. So don’t assume your teenager’s ChatGPT account has any of these protections in place unless you’ve specifically set them up together.
The honest bottom line
I don’t think AI is the enemy. My son is probably going to use these tools for the rest of his life, and I want him to use them well — with skepticism, with intention, and with the understanding that a chatbot is not a substitute for actually thinking things through himself.
Getting there is a process, not a single conversation. You’ll revisit it as he gets older, as the technology changes, as trust builds. That’s just parenting in this era.
If you want a tool that helps you manage the practical side — what apps your kid can access, how much time they’re spending on screens, where they are when you’re not with them — SmartDaddy is worth trying. There’s a 14-day free trial, no credit card required, and it works across both iPhone and Android.
