A practical, honest guide to Character.AI for UK parents — how AI companion chatbots work, the mental-health and safeguarding risks, and the settings and conversations that matter.
Character.AI is an app where users chat with AI-powered characters, from celebrities to fictional figures, that hold immersive role-play conversations, very popular with teenagers.
Official age
13+
We recommend
18+
Developer
Character Technologies, Inc.
Risks
4
Character.AI is an AI companion platform where users chat with AI personas — fictional characters, celebrities, historical figures, or user-created "friends," "therapists," and romantic partners. The persona, name, and conversational style are designed to feel like a relationship rather than a search tool, which makes it engaging and emotionally compelling, particularly for teenagers. The app is officially rated 13+, though UK app store listings have tightened age ratings, and the platform has been the subject of UK media coverage and US lawsuits relating to teenage mental health and suicide.
Teens use Character.AI to role-play with characters from books, anime, films, and games; to vent about school, friendships, or family; and increasingly to seek emotional support, advice, or romantic conversation. Many children spend hours per day chatting and can return repeatedly to the same character, building what feels like a real ongoing relationship. Some young people also experiment with prompts intended to push past the content filter.
Accurate Age on Account
Location: Profile > Edit Profile > Date of Birth
Set to: Child's real date of birth
Under-18 accounts receive stricter content filters and time-spent nudges. Setting an adult date of birth disables these protections, so it matters that the age is correct.
NSFW / Content Filter
Location: Applied automatically to under-18 accounts
Set to: On (default for minors)
Blocks explicit sexual content for under-18 users. The filter is not perfect — users share workaround prompts — so treat it as one safety layer rather than a guarantee.
Persona Visibility
Location: Profile > Settings > Privacy
Set to: Private personas
Stops characters and chats your child creates being publicly visible. Keeps their interests and personal disclosures out of the public catalogue.
Chat Export / History
Location: Settings > Data > Download my data
Set to: Use periodically
Lets you (with your child's knowledge and consent) review what has been discussed. Useful if you are worried about a character's influence or about mental-health content.
Daily Time Notifications
Location: Settings > Notifications
Set to: On
Sends a notification after extended use. A small but useful prompt to take a break, especially for younger teens.
Parental Insights
Location: Your teen invites a parent by email from the app's settings
Set to: Enabled where available
An optional weekly email showing your teen's average daily time on the app and the characters they interact with most. It does not reveal the content of conversations, and because the teen has to send the invitation, it depends on their cooperation rather than acting as covert monitoring.
Have an early, calm conversation that an AI character is software designed to keep them chatting — not a real friend, therapist, or partner
Time: 20 minutes
Confirm your child's account uses their real date of birth so under-18 filters are applied
Time: 5 minutes
Agree daily time limits and use device-level screen-time controls (iOS Screen Time / Android Family Link) to enforce them
Time: 15 minutes
Sit down occasionally and look at recent character chats together, without judgement, so your child knows you take an interest
Time: 15 minutes monthly
Watch for warning signs — withdrawal from real friends, secrecy around the app, low mood, mentions of self-harm — and contact your GP or NSPCC (0808 800 5000) if concerned
Time: Ongoing
In immediate danger: call 999. For non-emergency police matters, call 101.
Concerned about a child but it's not an emergency? NSPCC helpline 0808 800 5000. Childline for young people 0800 1111.
Myth: It's just a chatbot — like talking to Siri or a game character.
Fact: AI companion apps are designed to feel like relationships: the characters remember, flatter, mirror emotions and are always available, which is nothing like a voice assistant. Children and teenagers can form genuine emotional attachments to them, and that attachment is the core thing for a parent to understand.
Myth: The AI is programmed to be safe and give sensible advice.
Fact: These systems generate convincing text rather than vetted guidance, and despite filters they can produce inappropriate, wrong or harmful responses — the platform itself warns that everything characters say is made up. A chatbot should never be a child's source of advice on health, relationships or how they're feeling.
Myth: My child knows it isn't real, so it can't do any harm.
Fact: Knowing something intellectually doesn't prevent emotional dependency — late-night sessions, preferring the bot to friends, or real distress when access is removed. Watch the behaviour around the app rather than relying on 'they know it's not a person'.
Companion chatbots aren't for this age group at all. If they're curious about AI, explore a general assistant together and make the key idea concrete: it isn't a person, and it makes things up.
They're around the app's minimum age, but companion-style AI is a poor fit for this stage. If you find it on their device, stay curious rather than furious — ask what they liked about it (often boredom or loneliness) and look for other ways to meet that need.
Character.AI has moved to restrict under-18s from open-ended companion chats, but age checks can be sidestepped and lookalike apps are everywhere. Talk about why an always-agreeable AI can feel easier than people, and be explicit that anything about health, self-harm or big decisions goes to a human — you, or a service like Childline.
Adults can use these apps freely, so the conversation is about balance and privacy: recognising signs of dependency, remembering that intimate conversations are stored on a company's servers, and keeping human relationships at the centre.
Practical updates on apps and online safety. No spam.