How are the scenes and questions written?
Each moment is written by AI — based on who’s in the photo, the place, the light, what’s happening. Then it asks a question specifically for you.
Do you look through all my photos?
Only during the time period you selected. The scan runs on your phone — that’s where faces are found and grouped, and your library itself is never uploaded.
Two small things do leave: face crops, so Claude can double-check the grouping, and the few photos in the running for a moment. Both go over an encrypted connection to our own server, which hands them to Claude.
Your moment photos and the faces of your people are stored on a secure cloud server, so that we can print your book. Anything you delete is deleted from our servers too.
We never sell your data — not your photos, not your answers, not to advertisers, not to anyone. When you delete something, we delete it too.
How is AI used, exactly?
In a couple of ways — and your data is never sold, and never used to train AI.
- Face scanning — finding faces and grouping them, so your people can be followed through the year. This runs on your phone; only a few crops go to Claude to confirm the groups.
- The moments — Claude (Anthropic’s AI) looks at a handful of candidate photos, picks the ones worth writing about, and writes each scene and its question.
What if I fall behind?
A moment waits as long as it needs to, and the next one arrives on your usual day.
What does it cost?
Found & Felt is in early testing with a small group of families, free while we learn. Downloading a moment as a PDF is always free. Printing isn’t open yet — when it is, the price will be shown before you order, and nothing goes to press without you saying so.