AI Statement

Our Approach to AI

At Cherished Book, we recognise the profound sensitivity required when handling memorials and memories of loved ones. We use Artificial Intelligence (AI) thoughtfully to assist in the creation of the platform and to provide supportive features, but always with strict human oversight and strict data privacy.

In-House, Private Models

To ensure your data remains completely private, we use an in-house, open-source AI model. This means our AI runs entirely on our own secure servers. Your private data, memorial content, and personal information never leave our system and are never sent to third-party commercial AI providers.

There is one situation where content can reach another company, and it only happens if you ask for it. See If you connect your own AI assistant below.

If You Connect Your Own AI Assistant

You can choose to connect an outside AI assistant, such as Claude or ChatGPT, and ask it to create and look after memorials for you. This is entirely optional. Nothing below applies unless you turn it on yourself.

When you do, your assistant reads your memorial using the permission you granted it. We do not send anything to that company. Your assistant collects it. Where the reading happens depends on which assistant you chose, so if it runs on another company's servers, your memorial content is read there and is covered by their privacy terms rather than ours.

Two things are worth knowing before you connect one:

• Tributes are included. If you ask your assistant to help you read or moderate what people have written, it sees those tributes and the email address each person left. Those people wrote to you, not to an AI company, so please think about that before asking an assistant to go through them.
• You can stop it at any time. Open Connected apps in your account to see what you have connected, when it was last used, and disconnect it. We also email you the first time a newly connected assistant is used, so you find out quickly if it was not you.

A connected assistant can never email anyone, see memorials belonging to anyone else, publish something our safety check has refused, or change your account. You can read more on our AI assistants page.

How We Use AI

• Content Moderation: We use our private AI model to help filter out inappropriate content or spam, keeping memorials safe and respectful spaces.
• Memorial Content Generation: For families who need help finding the right words, our AI can offer supportive suggestions and help draft memorial text during the creation process.
• Platform Development: We use AI to assist in writing code and building features for Cherished Book.

Human Oversight and Review

We do not hand over control to machines. Every piece of functionality built with AI, and all creative content generated by the AI on our platform, is reviewed and checked by a human. We are committed to ensuring all content is sensitive, respectful, and appropriate for the context of memorialisation.

When AI Gets It Wrong

We acknowledge that AI is not perfect and can sometimes make mistakes, lack nuance, or misunderstand context. If you encounter any AI-generated content or moderation decisions that feel inappropriate, incorrect, or insensitive, we warmly welcome your feedback so we can improve our systems.

Please let us know by reaching out via our Contact Form.

Environmental Impact

Running AI models requires significant computational power. We are fully transparent about the energy footprint of our servers, including the AI processing. The energy usage of our AI systems is accounted for and offset as part of our broader environmental commitments.

You can read more about how our origin servers run on solar power in our Environmental Report.