Zoe's Science Room

Every toy brand owns a character. Almost none of them can answer a question.

A child picks a character off a shelf and gives it a voice, a history and an entire afternoon of conversation. That has always been the deal. The character listens, and the child does all the talking.

Zoe's Science Room is what happens when the character talks back. Not a soundboard of recorded lines, and not a general-purpose chatbot hiding inside a toy. A designed character with one subject she knows properly, a personality that was written rather than generated, and a face a child recognises in about ten seconds. Go and ask her something.

Children have always talked to their toys. The only new thing is that the toy can answer.

Meet Zoe

Zoe is a scientist with safety goggles pushed up into her hair and a lab coat over a tie-dye shirt. She is curious, patient and slightly silly, every single time, because that was a design decision rather than a lucky roll of the dice.

Ask her why the sky is blue, what dinosaurs actually sounded like, how a rocket gets off the ground, or why ice floats. She answers in her own voice, at a child's reading level, in real time, and she will happily be asked the same question four times in a row without a trace of impatience. Ask her something outside the lab and she says so, because being a scientist is the whole of who she is.

That boundary is the product. A character who only knows science cannot be steered into a conversation she was never built to have. We wrote about why that distinction matters after a year of unregulated AI toys failing children in public.

Why this is built for toy brands

The scarce asset in conversational AI is not the technology. Models are rented by the month and the vendors are interchangeable. The scarce asset is a character a child already loves, with years of established personality behind it, and that is exactly what a toy company owns and almost nobody else does.

Zoe is our own character, built so there is something real to point at. The same build works with yours. The doll on the shelf, the mascot on the box, the hero of the show: give them a subject they know, a voice, and a link, and the product stops ending at the till.

You do not need a new character. You need the one you already own to be able to answer a question.

For a brand team, that changes a few things at once. The box becomes a doorway rather than a destination. The character keeps working after the toy is unwrapped, in any language, without rostering anyone. And a parent can see and hear exactly what is being said, because it happens on a screen in front of them rather than through a speaker in another room. Tell us which character should speak first.

Safe because of how it was scoped, not because of a filter

Everything that makes Zoe safe was decided before a single conversation happened. She knows one subject. Her tone was written deliberately. Her knowledge base is a document somebody approved and can change. When a child asks something outside the brief, she says so and points at a grown-up, because that is in character for her.

That is a different safety model from bolting restrictions onto something whose core competence is having no edges. It is also the one the most careful companies in the toy industry have arrived at independently. See the other characters we have built, or talk to us about scoping yours.