← ForeverNear

ForeverNear — Data Handling

This is a plain-language summary of what data ForeverNear collects, where it lives, how long we keep it, and how you can have it erased. It is written for the families we serve and the funeral homes who host our presences. We would rather tell you the honest, in-progress truth than make a promise the product does not yet keep.

What we collect

We collect the materials a family gives us so we can build a presence of their loved one. This includes photos, video, and audio of the person, and voice samples taken from that media. It includes written intake notes about who they were — their stories, sayings, relationships, and the things they would and would not want said. It includes the family's own contact details (name, relationship, phone, and email) and the signed consent attestation captured during intake. Once a presence is live, it includes the transcripts of conversations visitors have with it, and a record of any crisis event flagged by our safety system (for example a visitor expressing self-harm, a medical emergency, or abuse) so a human can follow up.

Where it is stored

Structured records — the memorial's facts, consent attestations, conversation transcripts, and crisis events — are stored in Google Cloud Firestore. Uploaded files — photos, video, audio, and voice samples — are stored in Firebase Storage on Google Cloud. A hosted kiosk keeps a small amount of data on its local disk, including idle-clip frames and, on some deployments, a local crisis log. We rely on a few third parties to make a presence work, and some data is sent to them:

We do not sell your data, and we do not use it to train models for anyone else.

Consent

We do not build a presence from anyone without explicit, recorded consent. An authorized next of kin (or another person with clear authority, such as an executor or spouse) must sign an attestation that specifically covers using AI to recreate the person's voice and likeness. That voice-clone and AI-recreation consent is captured as its own attested field during intake, is reviewed by our team before any voice is cloned, and voice cloning is blocked in the system until that consent is approved.

How long we keep it

These are our policy targets for the pilot:

The source materials a family gives us are kept for as long as the memorial is active, and a family can request a full export of them at any time.

How deletion works

When a family asks us to forget a memorial, the presence is taken offline immediately and a 14-day grace window begins. Nothing is erased during that window, because we know a decision made in grief sometimes needs room to change. Until the window ends, the family can restore the memorial from their dashboard; a restored memorial goes back through our normal consent and activation review before it is live again. When the 14 days have passed, we remove the memorial from every place it lives. This erases the memorial's records in Firestore, its uploaded files in Firebase Storage, its cloned voice, its local artifacts on the kiosk, and the crisis-log lines tied to that persona. That erasure is permanent: once a memorial is erased it cannot be recovered. A family that needs everything gone at once can contact us, and an operator will carry out an immediate, permanent erasure with no grace window.

Security

Access to source recordings and structured records is restricted to a small trust team and the people a family names. Encryption-at-rest is in progress and is being rolled out; we do not yet claim it is fully live everywhere, and we will update this document when it is. We will state our shutdown commitments and export guarantees in each customer's written agreement.

Contact for data requests

To request an export, ask a question about this document, or ask us to forget a memorial, contact us at:

privacy@staim.ai