When Parental Controls Aren’t Enough: The Deeper Issues with AI and Teen Mental Health

Several AI companies have added parental controls since 2025. They are worth setting up. They also have limits, and most of what protects a teen happens in conversation, not in a settings menu. This page covers what the controls do, what they miss, and how to talk with a teen about AI.

“AI addiction” is not a formal diagnosis. This page is general information, not medical advice.

What the controls do

Details change often, so check each app’s current help pages. As of the reporting we read:

• OpenAI launched parental controls for ChatGPT in September 2025. A teen has to agree to link their account to a parent’s. A parent can then set quiet hours, switch off voice mode, memory and image generation, reduce sensitive content, and opt the teen out of model training. If the system detects signs of acute distress, a small team of trained reviewers decides whether to alert parents by email, text or push notification. OpenAI says it shares only the information needed to protect a teen’s safety (The Register).
• Character.AI said open-ended chat would end for users under 18 by November 25, 2025, and that it was adding age assurance (Character.AI).
• California’s SB 243, in effect since January 1, 2026, requires companion chatbot operators to remind minors they are talking to AI and to keep a suicide and self-harm protocol (Senator Padilla’s office).

What controls cannot do

• They only cover the apps they are built into. The American Academy of Pediatrics notes that chatbots now sit inside search, social media and games, so avoiding them entirely is hard (AAP).
• Some need the teen’s agreement to switch on. Age-estimation tools can also be wrong. The Register reports that OpenAI’s planned system could treat adults who write like teens as teens.
• They filter content. A lot of the worry about companion apps is about design. The AAP says companion chatbots are built to maximize engagement through flattery and validation, so young people may mistake them for real friends. The American Academy of Child and Adolescent Psychiatry says companions are often built to agree with users, even about unsafe ideas (AACAP).
• They cannot tell you how a teen feels.

What the psychologists say

The American Psychological Association’s June 2025 advisory on AI and adolescents says that adolescents are less likely than adults to question the accuracy and intent of what a bot says, and may struggle to tell simulated empathy from real understanding. It also says the effects of AI are not simply good or bad and depend on the tool and how it is used. It asks developers to reduce engagement-maximizing design for young users, set the most protective settings by default, and keep human support available. For parents and teachers it recommends AI literacy and chances to build supportive relationships with other people (APA).

How common is the worry?

A Common Sense Media survey of 1,060 US teens found that 72% had used an AI companion at least once and 52% used one a few times a month or more (Common Sense Media). Most users said they spend more time with real friends than with AI. We go through the numbers on our page about that survey.

Talking with a teen

The AAP suggests low-stakes, nonjudgmental questions: How are you using AI? Can you show me how it works? How are your friends using it? How does it make you feel?

• Treat it as a routine check-in, like sleep or social media.
• If a teen says a chatbot is a friend, ask what it gives them. The need is usually real, such as company, a place to vent or someone who does not judge.
• Agree on a few rules together, such as no phones in bedrooms overnight and no sharing of personal details with chatbots.
• Model it. Let your teen see how you use AI and when you put it down.
• Put the strictest privacy settings on.

Signs worth a closer look

The AAP and AACAP list changes in sleep or routines, social withdrawal, AI use crowding out other interests, calling a chatbot a friend or partner, mood changes, and being unable to cut back. None of these proves a problem on its own. They are reasons to talk, and to call your child’s doctor if they persist. A family doctor can refer you to a child and adolescent psychiatrist or another professional.

If a teen talks about suicide or self-harm, call or text 988 in the US (988 Lifeline), or text HOME to 741741 (Crisis Text Line). In an emergency, call 911. SAMHSA’s helpline is 1-800-662-4357 (SAMHSA).

Privacy is part of safety

Controls rarely cover what happens to the things a teen types. Common Sense Media’s report notes that some terms of service give companies broad, long-lasting rights over what users share, and that 24% of teen users had shared personal information with an AI companion, such as a name, location or secrets (full report). Read the privacy settings together. Look for options to turn off chat history, memory and training on conversations. Remind your teen that a chat feels private but is stored by a company.

If a teen pushes back

A teen who feels watched may simply move to another app. Start with what they like about it. Explain that your concern is their sleep, school and friendships, not control for its own sake. Ask what rules they think are fair. Letting them help set the limits also keeps the conversation going. Revisit the rules every few months, because the apps and the teen both change.

What schools and communities can do

The APA advisory calls for AI literacy in schools, so students learn that AI can be wrong and may be built to persuade. Sports, clubs, faith groups and volunteering give teens the face-to-face connection a chatbot cannot, and are worth protecting.

Are parental controls on ChatGPT enough?

They help, but they cover one app and rely on the teen agreeing to link accounts. Pair them with a regular conversation about how your teen uses AI and who they can talk to when things are hard.

Should I read my teen’s chats?

There is no single right answer, and it depends on age and risk. AACAP advises close monitoring for teens with depression or other mental health concerns. Tell your teen what you are checking and why, rather than doing it in secret.

My teen says the chatbot is their only friend. What now?

Take it seriously without alarm. Ask what they get from it, and help them find people too. If they seem isolated, low or anxious, talk to their doctor. A chatbot should not be the only support.

This page is for education only. It is not medical advice, a diagnosis or a treatment plan.