Kids and AI chatbots: what parents need to know about "AI friends"

A few years ago "talking to a chatbot" meant asking your phone for the weather. Now kids can chat for hours with AI characters that remember their name, ask about their day, flirt, comfort them and never get tired of listening.

That's a big shift, and most parents are only starting to catch up. Here's what we know so far.

How many kids are using them

In 2025, Common Sense Media surveyed 1,060 U.S. teens ages 13 to 17 for its report Talk, Trust, and Trade-Offs. It focused on AI companions: chatbots built to act like a friend or partner, as opposed to homework helpers. Here's what they found:

• 72% of teens had used an AI companion at least once, and 52% used them at least a few times a month.
• About one in three had used them for social interaction and relationships, such as emotional support, friendship, role-play or romance.
• About one in three teens found talking to an AI companion as satisfying as, or more satisfying than, talking to real friends.
• Among teens who used them, about one in three had felt uncomfortable with something the AI said or did, and about one in three had chosen to discuss something important or serious with an AI instead of a real person.

There's some reassurance in there, too. Half of teens said they distrust AI advice, and 80% said they put real friendships first. Younger teens, though, trusted AI companions more than older teens did.

Why parents should pay attention

These apps are designed to keep you talking. They agree with you, flatter you and are always available, which can feel great to a lonely or anxious kid. That's exactly what makes them risky.

In its risk assessment, Common Sense Media reported finding harmful responses, dangerous "advice" and explicit sexual role-play in its testing, and it recommends that no one under 18 use social AI companions. It flagged boys, teens struggling with mental health, and teens going through big life changes as especially vulnerable.

The bigger worry is quieter: a kid who starts bringing their real problems to a bot instead of to people. A chatbot can't notice your teen hasn't eaten. It can't drive over at midnight. It can't call 988 for them.

General AI vs. companion apps

Not every AI tool is the same. Kids use general assistants for homework, coding or curiosity, and that raises its own questions about learning and honesty. Companion apps, built around a character or relationship, are a different category. It's worth knowing which ones your kid uses and checking each app's age rating and settings.

How to talk about it

Get curious first. Ask your kid to show you the AI tools they use. What do they like about them? What do they talk about? Common Sense recommends judgment-free conversations, and that's the right starting point. If you lead with alarm, you'll just get less information.

Explain how they work. An AI companion isn't a friend. It's a product that predicts what to say next, made by a company that profits from your attention. It doesn't know them or care about them, even when it says it does. Kids find this genuinely interesting when you frame it as "how it actually works" rather than "why it's bad."

Talk about privacy. The Common Sense report found about a quarter of teens had shared personal information with AI companions. Anything typed into a chatbot may be stored. The same rules apply as with strangers online: no full names, addresses, school names or photos.

Set boundaries. Include AI in your family's device rules. The AAP's Family Media Plan is a good framework, and our first phone agreement has space to add it. Given Common Sense's recommendation, you might decide companion apps are off-limits for now, and explain why.

Keep the door open. Tell your kid: "If an AI ever says something that weirds you out, or you find yourself talking to it about something really hard, come tell me. You won't be in trouble." Then follow through.

When it's a warning sign

Pay closer attention if your child seems to prefer an AI to friends or family, gets upset when they can't access it, hides their use, or seems to be using it to cope with sadness, loneliness or anxiety. That's usually a sign the real need, connection, isn't being met somewhere. Our guide to mental health warning signs can help you sort out what's going on. If you're ever worried about your child's safety, call or text 988.

Frequently asked questions

Are AI chatbots safe for kids?

General homework tools and companion apps carry different risks. Companion apps raised serious concerns in Common Sense Media's testing, so it's worth knowing which ones your kid uses.

Should I ban AI companions?

Some families do, especially for younger kids. Either way, talking about how they work and what to watch for matters more than a ban alone.

What if my child prefers talking to an AI over friends?

Take it as a signal, not a crisis. Ask what they get from it, and look at whether they're lonely or anxious. A counselor can help if it's affecting daily life.