Clear boundaries make better conversations.

Key facts

ChatToAnyone safety at a glance

Each value below was checked against the product code, the linked page or the official site on the date shown. Report anything that no longer matches the app to safety@chattoanyone.com.

In detail

What happens to a message after you press send?

Your message appears on your screen at once, but it reaches the other person only after five steps: rate and size limits, a session check, a fast local screen for the most serious categories, an automated moderation model, and delivery. A flagged message is not delivered, an outage holds it back, and the path is identical on Free and Plus.

1. You press send. Your own copy appears on your screen straight away. The other person does not receive it yet.
2. Basic limits are applied. Each conversation has a sending-rate limit and a maximum message size, so a flood of messages or a giant paste is dropped rather than relayed. The service also rechecks that your session is still valid before anything is delivered.
3. The message is screened. A fast local screen looks for the most serious categories, such as sexual content involving minors. The message is then classified by an automated moderation model. If the check flags the message, it is not delivered and you see a notice that it may go against our community guidelines.
4. The check fails closed. If the moderation service cannot return a verdict, because of an outage or a timeout, the message is held back and you are asked to resend it later. An outage is not treated as misconduct, but it is never treated as permission either.
5. Only then is it relayed. A message that passes is delivered to the other participant and stored with the conversation.

AI companion replies go through the same kind of check before you see them, which is one reason an AI reply is not streamed word by word: the full reply has to pass moderation first. A companion is also instructed to say plainly that it is an AI whenever it is asked, and the “AI” label on the conversation is applied by the server, not by the model, so no prompt can remove it.

Automated moderation is a strong first layer, not a guarantee. Classifiers miss things, especially coded language and slow-building manipulation that looks harmless one message at a time. That is why the in-conversation controls below matter, and why this page asks you to keep personal details out of every conversation even though messages are screened.

Report and block

What happens when you report or block someone?

Reporting records the conversation, your chosen reason and a snapshot of the most recent message from the other side, then places it in a moderator review queue where content can be removed and accounts restricted or banned. Blocking asks for a second tap, ends the conversation and keeps you apart in matching until you unblock. Both controls sit inside the conversation itself.

Report this conversation

Choose the reason that fits best: Inappropriate, Harassment, Sexual content, Child safety, Spam or Other. The report records the conversation it came from, the reason and a snapshot of the most recent message from the other side, and it enters a moderator review queue as an open report. Reported content is kept for as long as the review and any legal duty require, even if a participant later deletes their copy of the chat. Conversations with an open report are excluded from automatic clean-up until the report is resolved.

Moderators can remove content, restrict or ban an account, and preserve evidence when the law requires it. Text that looks like child sexual abuse material is preserved as evidence automatically, whether or not anyone reports it, and confirmed material is reported to the relevant authorities. That is a legal duty, not a policy choice.

Block and leave

“Block and leave” asks you to tap a second time to confirm, then ends the conversation. A block works in both directions for matching: neither of you will be placed in a room with the other again, and a friend connection between you cannot be used to reconnect. You can review the people you have blocked under Settings, “Blocked people”. Unblocking does not restore a friendship.

When to use which

• Someone is rude or boring: just leave. Nobody is told why.
• Someone keeps pushing after you said no, or makes you uncomfortable: block and leave.
• Someone threatens you, sends sexual content, asks for money, or appears to be a minor in a human chat: report first, then block.

Recognize the pattern

What are the main risks in text chat with strangers?

Text chat removes the camera, which removes the most common way people are exposed in random video chat, but it does not remove manipulation. The five patterns worth recognizing early are requests for money, pressure to move to another app, sexual pressure or sextortion, a person who seems to be a minor, and someone in crisis. Each is covered below.

Requests for money, gift cards or crypto

A stranger who asks for money, a gift card code, cryptocurrency, “help with a fee” or your payment details is running a scam, however good the story is. The in-chat safety line says it plainly: never send money or gift cards to anyone you meet here. Report the conversation with the reason Spam or Other and block.

Moving the conversation somewhere else

Pressure to move to another app, especially early, is a common step in both scams and grooming because it takes the conversation away from moderation and reporting. Trying to move people off-platform to evade safety controls is against our terms. You do not owe anyone your handle.

Sexual pressure and sextortion

ChatToAnyone is safe for work and sexual content is not allowed. If someone pushes a conversation that way, leave. If anyone has obtained an intimate image of you and is threatening to share it, do not pay and do not send more. Keep the evidence, stop responding, and get help:

• If the image was taken when you were under 18, Take It Down, run by the National Center for Missing & Exploited Children, is a free service that helps remove or stop the online sharing of those images.
• If the image was taken when you were 18 or older, StopNCII.org, operated by the Revenge Porn Helpline (part of the charity SWGfL), creates a digital fingerprint of the image on your own device so participating companies can detect and remove copies. The image itself does not leave your device.
• Child sexual exploitation can be reported to NCMEC's CyberTipline.

Service descriptions checked on the official sites on 23 September 2026.

Someone who seems to be a minor

Human matching is for adults who declare that they are 18 or older. If a person in a human chat says or strongly suggests they are under 18, end the conversation and report it with the reason Child safety. Do not continue the conversation “to be nice”.

Someone in crisis

If the person you are talking to says they may hurt themselves or someone else, you are not expected to handle it alone. Encourage them to contact a crisis service, share Find A Helpline, and report the conversation so it reaches the review queue. If you believe someone is in immediate danger and you know where they are, contact emergency services.

Other chat sites

Using a random video chat site?

Our guides compare other random chat services, most of them built around video. The patterns above still apply there, and a camera adds a few of its own. This is the checklist our Omegle, OmeTV and Thundr guides send you to.

Before you turn the camera on

• Check the age rule. Most of the video services we cover say on their own pages that they are for adults 18 and over. If you are under 18, do not use them; ChatToAnyone offers a labeled AI companion from age 13 instead.
• Keep your surroundings out of shot. Anything on camera can be screen-recorded, including a school logo, a street sign or a letter on the desk.
• Find skip, block and report first. Look for them before your first conversation, while nothing is going wrong.
• Know what a purchase changes. Coins, gems or a Premium filter change who you are matched with. They do not make the conversation safer.

If something goes wrong on another site

End the call first, then report and block the account on that service and note the time and username. If someone recorded you and asks for money or more images, follow the sextortion steps above: do not pay, stop replying, keep the evidence and contact local police. In the United States, report anything involving a child to NCMEC’s CyberTipline; elsewhere, contact local police, and never download or forward the images, even to report them. If you feel shaken afterwards, the crisis services listed above can help.

Community guidelines

What are the community guidelines?

The community guidelines are ten rules that apply to every conversation on ChatToAnyone, with a person or an AI companion, on Free and Plus: respect people, keep it safe for work, never sexualize minors, no harassment or doxxing, no scams or impersonation, stay inside the safety net, photo rules, respect the AI label, nothing illegal, and no interference.

They build on the acceptable-use rules in the Chat Terms of Service and add a few of their own; where the two differ, the Terms govern. Moderation enforces many of them automatically, and moderators enforce the rest after a report.

1. Be the person you would want to be matched with

Disagree, joke, vent and ask awkward questions. Do not demean people for their race, ethnicity, religion, nationality, gender, sexual orientation, disability or any other part of who they are. A conversation can end at any time, and “no” ends a topic.

2. Keep it safe for work

No sexual or pornographic content, and no soliciting it from others, in text or photos. There is no adult mode and there will not be one. This includes “roleplay” framing and attempts to get an AI companion to produce sexual content.

3. Zero tolerance for anything that sexualizes minors

Sharing, requesting or producing child sexual abuse material, or sexualizing anyone under 18 in any way, results in removal and a report to the relevant authorities. Human matching is for adults only; users aged 13 to 17 may use labeled AI companions and nothing else.

4. No harassment, threats or doxxing

Do not threaten, stalk, intimidate or repeatedly target anyone. Do not post or ask for another person's private information, such as a real name, address, workplace, school, phone number or social profiles. Do not encourage anyone to harm themselves or others.

5. No scams, spam or impersonation

Do not ask for money, gift cards, crypto or payment details. Do not advertise, post referral links, or send the same message to conversation after conversation. Do not pretend to be another real person, a moderator, or ChatToAnyone staff.

6. Keep people inside the safety net

Do not pressure anyone to move to another app or share contact details to get around moderation or blocking. If a friendship forms, the in-app friend request exists so it can happen without either of you giving up private details.

7. Photos follow the same rules, and more

Where photo sharing is available, it only unlocks after ten messages in an eligible adult human conversation, and every photo is moderated before it can be shown. Do not send images of other people without their consent, and do not send anything that identifies a home, school, workplace, document or payment card.

8. Respect the AI label

Do not try to pass an AI companion off as a person, and do not try to trick a companion into claiming to be human or into bypassing its safety rules. AI replies can be wrong; do not treat them as medical, legal or financial advice.

9. Nothing illegal

Do not use ChatToAnyone to share illegal content, sell illegal goods or services, or plan or commit a crime.

10. Do not interfere with the service

Do not attack, overload, scrape or probe the service, automate conversations, or try to defeat moderation, age rules, regional rules or rate limits.

How the guidelines are enforced

• Message level: a message that fails moderation is not delivered, and the sender sees a notice.
• Conversation level: the other participant can leave, block or report at any time.
• Account level: after review, moderators may remove content, suspend or ban an account, and preserve evidence where the law requires it.
• Service level: if abuse outpaces review, the operators can switch person-to-person matching off so that everyone is routed to labeled AI companions until it is safe to reopen.

If you think we got a decision wrong, email safety@chattoanyone.com with the approximate date and time and what happened, and we will look at it again. Please do not include passwords, card numbers or identity documents.

For parents and guardians

What can a teenager do on ChatToAnyone?

A teenager aged 13 to 17 can talk only with clearly labeled AI companions, whose replies are moderated before display; there is no person-to-person matching, no invite code to a person and no photo sharing. ChatToAnyone asks every visitor for an age band before any conversation starts; that answer is a self-declaration bound to the browser, not an identity check.

A determined teenager can lie about their age on almost any website, including this one, so it is worth knowing what each answer unlocks.

• Under 13: the service stops there. There is no chat of any kind.
• 13 to 17: clearly labeled AI companions only. There is no person-to-person matching, no invite code to a person, and no photo sharing. Companions are instructed not to start or escalate romantic or sexual roleplay, and their replies are moderated before display.
• 18 and over: person-to-person matching where it is available, which is never in the United Kingdom and never when the visitor's country cannot be determined.

The most useful conversation to have with a teenager is about the patterns above, which look the same on every app: strangers asking to move apps, asking for pictures, offering money or gifts, or asking them to keep a secret.