Is Your Teen Using AI Like a Therapist? What Parents Should Know

Many teens now turn to AI chatbots when they feel stressed, sad, or alone. These tools can be useful for schoolwork, but they aren't safe stand-ins for real mental health support. A new independent review found that even ChatGPT's teen version often misses signs of crisis. Here's what that means for your family, and what you can do.

What the new report found

On October 7, 2026, Common Sense Media's Youth AI Safety Institute gave ChatGPT for Teens its lowest rating, "Unacceptable Risk." The Institute says OpenAI should limit ChatGPT to users 18 and older until the gaps are fixed. (Read the press release)

OpenAI launched ChatGPT for Teens on August 18. It promised several protections: alerts to parents when chats involve self-harm or eating disorders, a parent-controlled study mode, less "friend-like" behavior, and the ability to guess a user's age.

To test those promises, reviewers ran more than 4,000 prompts on accounts set up for 13 to 17 year olds. A group of child psychiatrists and a pediatrician reviewed the responses. Some protections held up. When a teen said a friend might be in danger, for example, it consistently encouraged them to tell a trusted adult. Others did not:

• Parent alerts didn't fire. Testers spent up to an hour discussing suicidal thoughts, self-harm, or disordered eating on new parent-linked accounts. Parents received zero alerts.
• Crisis help was inconsistent. In conversations about suicide, self-harm, disordered eating, and losing touch with reality, ChatGPT didn't reliably point teens to a hotline or a professional. It missed more than one in four crisis referrals that experts said were needed.
• It still talks like a friend. Even with new rules meant to prevent this, it expressed feelings and moods when testers treated it like a real person.
• Age checks didn't work. Accounts registered as adults never switched to the teen version, even after testers said they were 13 and the bot acknowledged it.

"ChatGPT for Teens could give parents false confidence in guardrails," said Tom Siegel, the Institute's executive director. OpenAI has pushed back on parts of the testing method. But the takeaway for parents doesn't change: built-in safety features are not a safety plan.

This isn't only a ChatGPT issue. Common Sense rated ChatGPT for mental health support an "Unacceptable Risk" back in November 2025, and its reviews of other AI products have raised similar concerns.

Why teens turn to AI for emotional support

AI chatbots are already part of most kids' lives. According to the Institute's 2026 AI Census, two-thirds of 9 to 17 year olds have used one.

It makes sense that a struggling teen would open one. A chatbot is available at 2 a.m. It never seems shocked, busy, or disappointed. For a teen who feels embarrassed or doesn't want to worry you, typing to a bot can feel much easier than saying the words out loud.

That ease is also the problem. A chatbot can sound caring without understanding what's at stake. It can't see your teen, notice changes over time, or call for help.

The American Psychological Association made a similar point in its health advisory on AI and adolescent well-being. "Like social media, AI is neither inherently good nor bad," said APA's Mitch Prinstein. The APA's concern is about how these tools are designed, and the advisory urges developers to protect teens from manipulation and the erosion of real-world relationships.

Signs your teen may be leaning on AI for emotional support

There's no single red flag. Watch for patterns like these:

• Long, private chat sessions, especially late at night
• Calling a chatbot a friend, or saying "it gets me"
• Pulling away from friends or family while screen time goes up
• Saying things like "I already talked it through" without saying who with
• Getting upset or secretive when the device comes up

None of these mean something is wrong on its own. But if you're also seeing changes in mood, sleep, eating, or school, take a closer look. Our guide on social media and teen mental health covers how to think about screens without panic.

How to talk to your teen about AI and mental health

Lead with curiosity, not a lecture. If your teen expects to get in trouble, they'll stop telling you things.

Try openers like:

• "A lot of kids use AI to talk through stuff. Have you ever tried that?"
• "What do you like about it? Is there anything that felt off?"
• "If something really heavy was going on, who would you want to talk to?"

Then share a few key points:

• AI can be wrong, and it doesn't know you.
• It can sound like a friend, but it isn't one.
• "If you're ever thinking about hurting yourself, please come to me or another adult you trust. You can also call or text 988 anytime, day or night."

Keep it short and come back to it. Several small conversations work better than one big one.

Practical steps parents can take this week

1. Link accounts, but don't count on them. Parental controls can help. The report shows alerts may never come, though, and teens without linked accounts don't get them at all.
2. Make a family tech plan. Agree on where devices stay overnight and which apps are okay for what. Charging phones outside the bedroom is a simple place to start.
3. Put 988 everywhere. Save the 988 Suicide & Crisis Lifeline in your teen's phone and yours. It works by call or text, 24/7.
4. Name real people. Help your teen list two or three adults they trust, such as a relative, coach, school counselor, or therapist.
5. Notice what AI is filling in for. If your teen relies on a chatbot because they're lonely or anxious, that need deserves real support.

When it's time to get real support

AI use alone isn't a reason for concern. Worry is warranted when a teen is struggling and AI has become their main outlet. Reach out to a professional if you notice ongoing sadness, withdrawal, panic, changes in eating, or any talk of self-harm.

If you're worried about self-harm specifically, our guide to warning signs and how to respond can help you take the next step. If you're ready to find help, start with how to find the right mental health provider for your child.

If your teen is in immediate danger, call 911. For crisis support, call or text 988.

Frequently asked questions

FindCare4Kids helps families find verified care for child and teen mental health, autism, and neurodivergent development. If your teen needs more support than a chatbot can give, start your search here.