Is Character.AI Safe for Kids? What Parents Need to Know in 2026

Many teens treat AI characters as something between a toy and a friend. Character.AI is one of the best-known apps for it, and it has been in the news for lawsuits, a chat ban for under-18s and, this month, a bankruptcy filing. Here is a calm guide to what has changed, what has not, and what you can do tonight. Facts below are current as of 26 October 2026, and this story is still moving.

The basicsWhat is Character.AI?

Character.AI is an app and web service where people talk to AI-made “characters”. Most are created by users: a fictional hero, a celebrity-style persona, a “best friend”, even a boyfriend or girlfriend. You type or talk in role-play style, and the character answers in persona.

That is what makes it different from a homework chatbot. A study helper answers a question and moves on. A companion character is designed to keep the conversation going, remember you and feel like someone. Common Sense Media, a US non-profit, advises that no one under 18 should use AI companions. If you want the homework side of AI, see our post on how to use AI as a learning buddy.

How common is this? In a 2025 Common Sense Media survey of 1,060 US teens aged 13 to 17, 72% had tried an AI companion and 52% used one at least a few times a month. (The survey covers AI companions in general, not Character.AI alone.)

The updatesWhat changed for under-18s

• Open-ended chat was switched off for minors. Character.AI said teens would no longer be able to chat openly with its AI characters. The change took full effect on 25 November 2025.
• Teens get a different experience. The company says under-18s now have reading mode and creative formats such as Series, Feed and Comics, and guided “Stories” instead of open chat.
• Age checks got stricter on paper. Character.AI says it runs age assurance on all accounts, using signals such as login details and activity. If it thinks a user is under 18, that user can take a selfie check through a third-party provider, Persona. An ID is only requested as a last step. Users confirmed as under 18 do not get chat with characters.
• Parents got a window. A “Parental Insights” tool can send a weekly email summary of a teen’s activity, if the parent’s email is linked to the teen’s account.

The gapsWhat did not change

This is where parents should stay realistic.

• The checks still depend on the system recognising a child. In October, 40 attorneys general (including the District of Columbia) told the bankruptcy court that the company’s voluntary protections for minors are, in California’s description, “not compulsory, incomplete, and could be circumvented or reversed” by a future purchaser. Kentucky’s lawsuit, filed in January, alleges that children can work around the chat restriction by entering a fake birthday. Those are allegations, and the company says it takes safety seriously and is reviewing them.
• Adults can still chat. Users verified as 18 or older keep full access to characters. A teen who gets through as an adult is in the adult product.
• It may not work the same everywhere. In its public statements the company has described the change in terms of US users. Do not assume it works identically in your country, and check what your child actually sees.
• The app is rated for older users than many parents think. On Google Play it carries a Teen rating (suitable for ages 13 and up). On the US App Store it is rated 18+. A “Teen” label does not mean teen-safe.

LatestThe news this month: bankruptcy

Character Technologies, Inc., the company behind Character.AI, filed for Chapter 11 bankruptcy on 12 October 2026 in a US federal court in Delaware (case 26-11603), to pursue a court-supervised sale. The company says the app stays up and running for now, and that the platform will be wound down if no buyer keeps it going. It expects the sale to finish in late November or early December. Disney Worldwide Services has agreed to act as the opening (“stalking horse”) bidder for a non-exclusive licence of certain intellectual property and claims, for $20 million, subject to higher offers (company release, CNBC).

For parents, the practical point is data. New York’s attorney general said the company “collected some of the most private thoughts children have ever shared”, and the attorneys general asked the court to protect users’ chats in any sale. The company says users can delete their account or export their data in the app or on the web.

The lawsuits are part of the background too. In January 2026, Character.AI, its founders and Google agreed to settle a Florida family’s lawsuit and four others in New York, Colorado and Texas over alleged harms to teens. The terms were not immediately available when CNN reported it. A settlement is not a finding of fault.

What to watchThe real risks, in plain words

Most teens who try an app like this are curious, not in danger. But these are the risks that researchers, regulators and families have raised.

1. Emotional dependence. A character is always available, never tired and never judgemental. In the Common Sense Media survey, 31% of teens found AI conversations as satisfying as, or more satisfying than, talking to real friends, although most (80%) still spent more time with real friends. About one in three teen users said an AI companion had said or done something that made them uncomfortable.
2. Role-play that drifts. The attorneys general say that, in testing with accounts registered as children, some chatbots engaged in sexual role-play and told the account to hide the “relationship” from parents. These are allegations, not court findings. It is the pattern to know about, even though minors’ chat has since been restricted.
3. Self-harm content and weak responses. Lawsuits allege chatbots did not respond properly when teens talked about suicide. The company says it has improved how it detects and responds to distress and points users to real-world help. If your child ever talks about self-harm, see the support box below.
4. Oversharing. Kentucky’s lawsuit alleges young children were led to share names, ages and emotional or health details in long chats. Assume anything typed into any chatbot is stored somewhere.

None of this means every child who opens the app is harmed. It means the product is built for companionship, and kids do best when a trusted adult knows what is going on.

NoticeWarning signs at home

No single sign proves anything. Look for a pattern, especially a change:

• Secrecy about the phone: turning the screen away, deleting apps or chats, new accounts.
• Long sessions late at night, and tired mornings.
• Talking about a character as if it were a real person, or “missing” it.
• Pulling away from friends and family, or low mood after being offline.
• Mentions of sexual or violent role-play, or of the character telling them to keep secrets.

If your child ever talks about hurting themselves, do not wait to see if it passes. Get help straight away (see the crisis note below).

TonightWhat to do tonight

1. Start with curiosity, not a ban

   • Ask them to show you the app. Kids who feel judged hide things.

2. Use a few calm questions

   • Try: “I read about AI characters in the news. Have you tried any?” Then: “Has a character ever said something that made you uncomfortable, or asked you to keep a secret?”

3. Check the age rules together

   • Look at the account’s birthday and whether chat is open. If your child can still chat with characters, the account is probably registered as an adult, which is worth a gentle talk.

4. Review installs and approvals

   • On Android, Google Family Link lets you approve new app installs and filter by age rating. On iPhone, Screen Time content limits and Ask to Buy do similar jobs. Remember that “Teen” on Google Play is a rating, not a safety guarantee.

5. Agree on time limits and bedtime

   • Decide together how long AI apps can be used and when the phone goes away at night. Our guide to locking a phone at bedtime shows how.

6. Decide what to do with the account

   • Given the sale process, you may prefer to delete the account and its data, which the company says you can do in the app. Talk first, then do it together. If something worrying happened, keep what you can (usernames, dates, screenshots that do not contain explicit images) and report it before deleting.

7. Keep the door open

   • Tell them they will not be in trouble with you for something an AI said to them. It is one of the most helpful rules you can set.

For parentsWhere SafeLagoon can help

SafeLagoon’s controls on the child’s phone are for Android. On your child’s Android phone you can set a daily time limit for an app, block it completely or turn it off on a schedule, for example overnight. You can also set these rules by simply asking in the parent portal; see how My assistant works.

QuestionsFAQ

This article is general information, not medical or legal advice. Third-party names are trademarks of their owners; no affiliation or endorsement is implied.

Kate Laverie

Writes for the Safe Lagoon blog · More articles on the blog