Machine learning
Hey everyone,
Im using a self hosted instance of ente. Can someone let me know if the machine learning works 100% locally? Are my files sent elsewhere or used for AI Learning?
I don’t know a tonne about how AI or ML actually works, but my understanding is it depends on a collective. So I don’t know how that would work on its own
3 Replies
Ente uses end-to-end encryption, so the server doesn't have access to your data. Your files are encrypted on your device before being uploaded, and only you have the keys to decrypt them. For more details, see the architecture document at https://ente.io/architecture.
It is beneficial to enable machine learning before importing your photos, as this allows the Ente app to index your files as they are uploaded. The indexes are synced across your devices using the same end-to-end encryption used for syncing your photos, as mentioned at https://help.ente.io/photos/features/machine-learning.
-# If your issue is resolved, feel free to use the /solved command to close this thread. If you'd like to ask me another question use /docsearch
Yes, it is 100% local. The app downloads prebuilt static models when it starts machine learning, but the processing all happens only on your device. The generated indexes are then end to end encrypted (like your photos etc) and stored so that your other devices can use them without needing to run the entire process again.
More details are here - https://ente.io/ml/
ente
Machine Learning
Learn how Ente uses on-device machine learning to power private, end-to-end encrypted features like face recognition and semantic search—without a cloud.