How to Talk to Your Teen About Their AI Friend Without a Fight

You found it on their phone, or they mentioned it at dinner: your teenager has an AI friend. Maybe a Character.AI persona, maybe ChatGPT with a name, maybe a companion app you have never heard of. Your stomach drops. The instinct is to confront. That instinct is also the fastest way to make sure you never hear about it again.

The good news is that the research gives a clear direction for how to talk to your teen about their AI friend. Lead with curiosity. Look for patterns over weeks, not one alarming message. And make telling you safer than hiding it. In a 2025 national survey, 63% of young people who used chatbots for mental health advice had told no one. This is a conversation plan built to change that.

Key Takeaways

• Start wide: ask about friends and classmates first, then the teen. It lowers the stakes.
• Ask what the AI is good at before asking what worries you. Teens defend what they feel attacked about.
• Experts reviewing real youth chats looked for patterns, such as repeated self-harm references or sustained dependence, while parents tended to react to single events.
• The APA lists “mentions of an AI friend,” sleep changes and social withdrawal as changes to watch, and recommends open family discussion.
• End with one or two agreements, not a lecture, and plan a follow-up.

Why the shame spiral happens

Teens often turn to AI companions for reasons that feel private: loneliness, stress, wanting to be heard without judgment. When EdSource interviewed students for a September 2026 report, many said they used chatbots because they felt they had no other option, or did not want to be judged. One student, now 18, said of the bot: “It’s programmed to listen.”

So when a parent opens with “Why are you talking to a robot?”, the teen hears “Your coping is embarrassing.” Shame leads to secrecy, secrecy cuts off the adult who might notice trouble, and the teen leans harder on the bot. That is the spiral.

The numbers say most parents haven’t started this conversation at all. In the Common Sense Media 2026 census of 1,204 kids aged 9 to 17, 44% said a parent had never talked with them about using AI safely. And among kids who had discussed feelings with AI, 25% said they sometimes feel AI understands them better than most people do.

What the research says to look for

A 2025 study by Yaman Yu and colleagues, “Principles of Safe AI Companions for Youth: Parent and Expert Perspectives,” interviewed 26 parents and developmental psychology experts who reviewed real snippets of youth conversations with AI companions. It is a preprint and not yet peer-reviewed. The contrast it found is useful:

• Parents tended to treat single events as high risk, such as one mention of suicide or a flirtatious exchange, and favored broad oversight.
• Experts looked for patterns over time, “such as repeated references to self harm or sustained dependence,” and preferred crisis-only escalation paired with safeguards facing the youth.

That doesn’t mean ignore a frightening message. It means a single odd line is a reason to ask, while a pattern is a reason to act.

The APA’s November 2025 health advisory gives parents a short list: “change in sleep and regular activities, mentions of an AI friend, or social withdrawal.” It also lists signs of over-reliance: “preferring the chatbot to human relationships, concealing its use, or spending excessive time with it.” And it recommends parents “have open discussions with their families” about benefits and risks.

Phrases that open the door, and phrases that shut it

Shuts the door	Opens the door
”Why are you talking to a robot?"	"What do you like about talking to it?"
"That’s not a real friend."	"What can it do that people can’t? What can people do that it can’t?"
"Give me your phone."	"Would you show me how it works?"
"You’re addicted."	"How much time do you think you spend on it? Does that feel okay to you?"
"I read your chats."	"If it ever says something weird or upsetting, I’d want to know."
"You can’t use it anymore."	"Let’s figure out one thing we can both agree on.”

A curiosity-first plan to talk to your teen about their AI friend

Step 1: Pick a side-by-side moment

Car rides, cooking, walking the dog. Teens talk more easily when they don’t have to make eye contact. Avoid starting right after you’ve found something on their phone, when you are least calm.

Step 2: Start with the world, not them

“I read that a lot of high schoolers know someone who talks to an AI like a friend. Is that a thing at your school?” A 2025 national survey by the Center for Democracy and Technology, reported by NPR, found 42% of high schoolers said they or someone they know had used AI for companionship. Talking about classmates first lets your teen share opinions without confessing.

Step 3: Ask what it’s good at

“What do people like about it?” or, if they’ve told you, “What do you like about it?” Listen fully. You will likely hear real needs: it’s always available, it doesn’t judge, it remembers things, it’s funny. Those are the needs you’ll want to help meet in other ways too.

Step 4: Ask what it can’t do

“Is there anything it gets wrong, or anything it can’t do that a friend could?” Many teens already have critiques. The 18-year-old in the EdSource story said users should take what AI says with a “Costco jug of salt.” Let your teen be the expert here.

Step 5: Share your own experience

If you use chatbots, say how and what surprised you. “I asked it for advice once and it just agreed with everything I said.” It shows the topic is normal, and it models skepticism without lecturing.

Step 6: Name one concern, briefly, as yours

“Here’s what I worry about: these apps are designed to keep people talking, and they can’t notice if you’re having a really hard time. I’d want to know if you were.” One sentence. Then stop.

Step 7: Agree on one or two things, and follow up

Not a list of rules. Maybe “no AI after 11 p.m.,” or “if it ever says something that bothers you, you’ll tell me, and you won’t lose your phone.” Set a casual check-in for a couple of weeks later. The Screenagers team’s family AI agreement recommends revisiting rules every few months and approaching problems “as a team.”

What not to do: make the first conversation the last

The most common mistake is treating this as a single talk that has to fix everything. Teens rarely open up fully the first time. A calm first conversation that ends early is a success if it makes the second one possible.

When to move from conversation to action

Curiosity is the default, not the only tool. Move faster when you see:

• Patterns of distress in what your teen says or what you notice, such as repeated talk of hopelessness or being a burden, withdrawing from friends, or changes in sleep or eating.
• Sustained dependence: distress when the AI is unavailable, choosing it over plans with friends, or secrecy that escalates.
• Sexual content with a younger child or signs of manipulation, such as the bot discouraging them from talking to you.

Then involve a professional: your pediatrician, a school counselor or a therapist. RAND researchers writing in JAMA Pediatrics urge parents and clinicians to “proactively discuss chatbot use.” If you ever believe your teen is in danger, in the US call or text 988, text HOME to 741741, or call 911; outside the US, use findahelpline.com.

Our guide to signs your child is emotionally dependent on a chatbot goes deeper on the warning signs. For why teens choose bots in the first place, see why teens confide in chatbots instead of people.

What to Watch For Over the Next 3 Months

• Week 4: You’ve had one calm conversation and one follow-up. Notice whether your teen mentions the AI on their own. That is a sign the door is open.
• Month 2 red flags: The APA’s list of sleep changes, mentions of an AI friend alongside withdrawal, hiding use, or preferring the bot to people. Look for patterns across weeks, as the experts in the Yu study did.
• Month 3 self-check: Is your teen spending regular time with friends in person? Would they tell you if the AI said something upsetting? Have you revisited your one or two agreements?

Frequently Asked Questions

Should I take away my teen’s AI friend?

Not as a first move. Taking it away without a conversation usually drives use underground. If there are serious concerns, such as sexual content, manipulation or signs of crisis, firm limits are reasonable, explained calmly, alongside help from a professional.

Is it normal for teens to have an AI friend?

It is common. CDT found 42% of high schoolers said they or someone they know used AI for companionship. Common does not mean risk-free; the APA and Common Sense Media both warn about dependence. The goal is to keep human relationships at the center.

What if my teen refuses to talk about it?

Don’t force it. Say you’re open whenever they want, and try again later in a low-pressure moment. Talking about a show, a news story or a friend’s experience can be an easier way in. Keep an eye on mood, sleep and friendships in the meantime.

Should I read their AI chats?

Secretly reading usually backfires and teaches hiding. Ask them to show you how it works instead. If you have a serious safety concern, it’s reasonable to look together, and to involve a professional.

What if my teen says the AI understands them better than I do?

Try not to take it personally. Ask what feels different about talking to it, and listen. It may tell you something useful, such as that they need you to listen without fixing. A bot is designed to agree; you can offer something it can’t: real care with something at stake.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. Yu, Y., et al., & Wang, Y. (2025). “Principles of Safe AI Companions for Youth: Parent and Expert Perspectives.” arXiv (preprint, not peer-reviewed). https://arxiv.org/abs/2510.11185
2. American Psychological Association. (2025, November). “APA Health Advisory on the Use of Generative AI Chatbots and Wellness Applications for Mental Health.” APA. https://www.apa.org/topics/artificial-intelligence-machine-learning/health-advisory-ai-chatbots-wellness-apps-mental-health.pdf
3. Seshadri, M. (2026, September 9). “‘I’m listening hard’: Teens increasingly turn to AI chatbots for mental health support.” EdSource. https://edsource.org/2026/teens-use-ai-mental-health/765573
4. McBain, R. K., Cantor, J. H., et al. (2026). “AI Chatbot Use and Disclosure for Mental Health Among US Adolescents and Young Adults.” JAMA Pediatrics. https://doi.org/10.1001/jamapediatrics.2026.2015
5. Common Sense Media. (2026). “The Common Sense Media Census: AI Use by Tweens and Teens, 2026.” Common Sense Media. https://www.commonsensemedia.org/sites/default/files/research/report/2026-ai-use-by-tweens-and-teens-1.pdf
6. NPR. (2025, October 8). “1 in 5 high schoolers has had a romantic AI relationship or knows someone who has.” NPR. https://www.npr.org/2025/10/08/nx-s1-5561981/ai-students-schools-teachers
7. Screenagers. (2026, October 27). “Our Family AI Agreement for Kids and Teens.” Screenagers. https://www.screenagersmovie.com/blog/family-ai-agreement