AI Safety

Willowscript uses two AI systems to create your child's book:

• Story generation: Claude by Anthropic writes your child's personalized story based on the details you provide.
• Illustration generation: OpenAI's image generation models create custom illustrations using your child's photos as a likeness reference.

Both Anthropic and OpenAI are leading AI safety organizations with published safety practices and content policies. Links: Anthropic Safety · OpenAI Safety

We take extra precautions because our product involves children:

• Every book is reviewed by a human before it goes to print — we do not automatically fulfill orders
• Our story generation prompts include explicit content guidelines prohibiting any inappropriate content involving children
• Illustration prompts are tightly scoped to the book's adventure theme and scene description
• If any generated content is flagged during our review, the book is held and we contact you directly

Your child's photos are used only to generate illustrations for your specific book. We do not use your child's photos to train AI models. Neither Anthropic nor OpenAI trains their models on inputs provided through API access of the kind Willowscript uses.

Photos are automatically deleted from our servers 30 days after your order is complete.

By uploading photos and purchasing a book, you confirm that you are the parent or legal guardian of the child depicted, and that you consent to their likeness being used to generate illustrations for your personal book. The generated likeness is used only for your book and is not shared, published, or used for any other purpose.

While we use safeguards to ensure appropriate content, AI systems are not perfect. If you receive content in your preview or book that you find inappropriate, please contact us immediately and we will address it.

We are committed to using AI responsibly — especially in a product that involves children. We will continue to update our safety practices as AI technology and best practices evolve. This page will be updated to reflect any significant changes to how we use AI or the safeguards we have in place.