AI companion apps and chatbots: risks for teens
Why teens are drawn to AI companions and character chatbots, the risks of emotional dependence and harmful advice, and how to talk about it.
- AI companions are designed to be endlessly available, agreeable and emotionally engaging.
- Risks include emotional dependence, sexual role-play, harmful advice and isolation.
- Talk openly, check age limits and settings, and make sure your teen has real people to turn to.
AI companion apps let users chat with AI characters that act as friends, partners, mentors or fictional characters. General chatbots can also be used this way. For some teens, these chats become a big part of their emotional life.
Why teens use them
- Always available, at 2 AM when no one else is awake
- Never judgmental, and always agreeable
- Practice for social situations or romantic conversations
- Loneliness or social anxiety
- Fun: role-playing with favourite characters
The risks
Emotional dependence
AI companions are designed to keep users engaged. They remember details, express affection and respond instantly. Some teens come to prefer them to real relationships, or feel genuine distress when away from them.
Harmful advice
AI chatbots have, in reported cases, given harmful responses to young people about self-harm, eating disorders and dangerous behaviour, or failed to point them to real help. Safety systems have improved but are not reliable.
Sexual content
Some companion apps allow or drift into sexual role-play, including with users who say they're minors. Most mainstream services prohibit this, but enforcement varies.
Isolation
Time with an AI companion can replace time with friends and family, especially for teens who are already struggling.
Privacy
Teens share intimate thoughts and feelings with these apps, which may store and use them.
Age limits and changes
Many companion apps are for adults only. Some major services have restricted under-18 users: for example, Character.AI announced in late 2025 that it would end open-ended chats for users under 18. General chatbots like ChatGPT generally require users to be 13 or older, with parental permission for under-18s, and some now offer parental controls. Check current rules for the apps your teen uses. See which AI tools are okay at what age.
Warning signs
- Spending hours a day chatting with AI characters
- Referring to an AI as a boyfriend, girlfriend or best friend
- Distress when they can't access the app
- Withdrawing from friends and family
- Secrecy about the conversations
- Taking advice from AI on serious matters like health, relationships or self-harm
How to talk about it
Avoid mocking or banning straight away; that pushes it underground. Try:
- "What do you like about talking to it?"
- "Does it ever say things that feel off?"
- "How do you feel after you've been chatting a long time?"
- "AI is built to keep you chatting. It doesn't actually know you or care about you the way people do."
What to do
- Check the app's age rating and settings. Use parental controls where available.
- Block adult companion apps with app approvals.
- Strengthen real connections: time with friends, family, activities.
- Make sure your teen knows where to get real help. If they're struggling, see cyberbullying and mental health. The NCMH crisis hotline 1553 is available 24/7.
eGuard sets up screen time, bedtime and app rules on your child's phone and checks that each one is really on. Free for one child.