Child safety standards

RateAI has zero tolerance for child sexual abuse and exploitation (CSAE). Content that sexualises a minor, and any behaviour aimed at exploiting one, is prohibited on this service without exception — no warning, no second account, no appeal on the merits of the content.

This page is the published standard behind that sentence: who the service is for, what is forbidden, what we do to keep it off the platform, how anyone can report what we missed, and who to write to. It applies to the app on every platform we ship to, and to this website.

These standards sit alongside our terms of service, which every account accepts, and are written to meet Google Play's child safety standards policy and the equivalent expectations of the App Store.

You must be 18 or older to hold an account. That is not a checkbox on a sign-up screen. A date of birth is required during onboarding, and the minimum is enforced in the database itself: a profile whose birth date is less than eighteen years old is refused at write time, whatever the client sends.

Nothing in RateAI is designed for, marketed to, or appealing to children. There are no child-directed features, no content aimed at minors, and the store listings are rated for an adult audience.

If we find that an account belongs to someone under 18 — from a report, from moderation, or from anything else — the account and its content are deleted. They are not warned and not restricted; they are removed.

The following are banned outright, regardless of intent, framing, or whether the depiction is a photograph, a drawing, or generated by a model:

• Child sexual abuse material (CSAM) in any form, and links or references pointing to it anywhere else.
• Sexualisation of a minor — imagery, text, or commentary that presents a person under 18 in a sexual way, including rating notes written about someone who is or appears to be a minor.
• Any photograph of a minor. Every photo on RateAI has to be of the account holder, and every account holder is an adult, so a child in an uploaded image is a violation on its own — no sexual element is required.
• Grooming, solicitation, or sexual extortion — using the service to contact, befriend, pressure, or blackmail a minor, or attempting to move such contact to another platform.
• Trafficking or advertising a minor for sexual purposes, or seeking such material from other users.
• Impersonating a minor, or presenting an adult profile as a child.

Breaking any of these ends the account permanently and, where the law requires it, produces a report to the authorities.

RateAI is built so that a photograph is checked before anyone else can see it, rather than taken down afterwards:

• Nothing is published on arrival. An uploaded photo is visible to nobody but its owner until it has been classified. It does not appear in the rating deck, in the games, or on a profile while it waits.
• Automated classification runs on every upload, and “depicts a minor” is one of the categories it decides. A photo it rejects is never shown to another user.
• Uncertainty goes to a person, not to the deck. An image the classifier flags, or one it cannot read at all, is held for human review. A failure to read a photograph is never treated as permission to publish it.
• A rejection sticks. The fingerprint of a rejected file is barred, so deleting the row and uploading the same image again produces another rejection rather than a fresh start.
• User-written text is reportable everywhere it appears. Rating notes are the one channel where a user's own words reach another person, and every one of them can be reported from the app.
• Blocking is immediate. Any user can block any other, which takes effect in the deck they are holding.

Moderation decisions are recorded with the account that made them, so every removal and every restoration is auditable after the fact.

Reporting is built into the app, wherever content appears. The overflow control on a photo in the rating deck reports that photo; the same control on a profile card reports the profile; a long press on a rating reports what was written; and the header action inside a game reports the person behind the face on screen. Choose “Appears to be a minor” from the list of reasons.

You do not need an account, and you do not need to be the person affected. Anyone can write to [email protected] — put child safety in the subject line and it is treated as urgent. The contact form reaches the same inbox.

Please do not attach, forward or re-share suspected abuse material. Send us the profile name, the screen you saw it on, and the time. We can find the content from that, and passing the material around multiplies the harm.

Reporting to us does not replace reporting to the authorities, and you should never feel you have to wait for us. A child in immediate danger is an emergency call (112 in the EU, 911 in the US). Abuse material can be reported directly to NCMEC's CyberTipline, to your national INHOPE hotline, or to the police.

Reports involving a minor are prioritised ahead of everything else in the review queue and are looked at by a person, not closed automatically.

Where a report is confirmed:

• The content is removed from view immediately, and its fingerprint is barred from being uploaded again.
• The account is terminated permanently — not suspended — along with any other account we can attribute to the same person.
• The material and the account records are preserved in secured storage for the authorities, rather than deleted with the account and rather than forwarded to anyone by email.
• Confirmed child sexual abuse material is reported to the competent authorities. RateAI is operated from Germany, so that means the Bundeskriminalamt (BKA); we also file with NCMEC's CyberTipline, which accepts reports from providers outside the United States.
• We respond to lawful requests from law enforcement, and cooperate with the platforms we ship through where they ask.

Removals are reversible, and if we take down a photo we should not have, writing to [email protected] gets it restored with the reason cleared. That does not extend to CSAE: an account ended for it stays ended.

The designated point of contact for child safety matters — including enquiries from Google, Apple, law enforcement, and child protection organisations — is the operator of the service:

Jan Tabens
RateAI — child safety point of contact
[email protected]

Postal address and provider identification are on the imprint. Messages about child safety are answered ahead of ordinary support, and we aim to acknowledge them within one working day.

These standards are reviewed at least once a year, and whenever the product changes in a way that touches them — a new surface that shows other people's photographs, or a change to how uploads are moderated. The date at the top of this page is the current version.

Questions about this page: [email protected].