Children and family safety

Our position, plainly

Generative AI is a powerful tool with real failure modes. We do not think the right response is to hand it to a six-year-old with a cartoon dragon on the launch icon and hope for the best.

So Doodle Dragon Studio is built as a parent-operated studio. The adult holds the account, drives the AI tools, and reviews the working project. The child does the part children are brilliant at: drawing, imagining, and telling you what happens next.

The books are for children. The controls are for grown-ups.

What that means in practice

The account holder must be 18 or over. You confirm this when you sign in, and AI creation is only available to a signed-in adult account holder.

We do not create child accounts. There is no child login and we never ask a child for a date of birth. A legacy local preferences screen can store a learner name and approximate age on the device; it is not used as a cloud identity and is scheduled for replacement by a real Accessibility screen.

A child’s name in a story is just text. If you type your daughter’s name into a page, that is content in your project on your device. It is not a profile, it is not identity data, and we do not build anything from it.

Every request is initiated and reviewed by you. The app shows the relevant prompt, settings, images, or frames before sending. Standalone illustrations and videos also have output review screens. Planner artwork selected during plan review and generated narration can be attached to the working project when they finish, so the adult must review the whole book before export or sharing.

Nothing is published. There is no public gallery, no feed, no sharing to strangers, no “discover” tab. What you make goes where you send it and nowhere else.

What we will not build

Being specific matters more than reassurance here:

• No face recognition and no biometric templates. Ever.
• No likeness transfer that reproduces a specific real person’s face.
• No automatic inference of family relationships.
• No public sharing, gallery, or social feed.
• No advertising, and no advertising SDKs.
• No selling or renting of personal information.
• No child accounts.

About safety filters

The AI providers we use apply their own content filters. They work, and they are also blunt: they refuse plenty of entirely innocent family content, sometimes for reasons that seem absurd.

When a request is refused, your credits come back automatically and you can rewrite and retry. See generation errors.

We would rather a filter be over-cautious than under-cautious, and we would rather tell you it happens than pretend it doesn’t.

Photographs of your children

If you import a photo of your own child, it stays on your device unless you ask for an AI action on it. If you do, the app strips the metadata first and shows you exactly what will be sent.

You should think about this before doing it, and the app is designed so that you can think about it — that is what the review step is for. Many people use the app entirely with drawings and never send a photo of a child anywhere. That is a completely reasonable way to use it.

Doing this with your child

A few things that work well:

• Let them draw first, without any talk of AI. The drawing is the point.
• Ask them what happens next rather than telling them.
• Read the description out loud before you generate. They will correct you, loudly and specifically, and the result will be better.
• Keep the original sketch on a page too. Seeing both is half the delight.
• Record them narrating it. In ten years that will be the part you care about.

Reporting a problem

If the app produces something inappropriate, we want to know immediately. Email support@mcntech.com with what you asked for and what came back. These reports go to the front of the queue.