Details on Privacy Mode with storage - how its different than Privacy Mode (legacy)

Discussions
1

Hey I’m trying to understand what the actual difference is for the new privacy mode announced on June 19th called “Privacy Mode with storage”.

Looking at the previous versions of the security page:

vs

Old version:

  • “We guarantee that code data is not stored in plaintext at our servers or by our subprocessors”

New version:

  • “We guarantee that code data is never stored by our model providers or used for training”

With this change there is quite a lot more trust bestowed from the user to Cursor - It would be helpful to understand the controls being applied for secure storage now that it’s happening & the reason to use this new mode versus staying on the legacy mode.

2 Likes
2

Hi @Aaron_Bacchi and welcome to Cursor Forum.

thank you for the detailed question.

The current “Privacy Mode” allows users to turn on Background Agent, Memories and User Rule synchronization while keeping code and API requests private so they are not used for training etc. For this features specific data (like memories…) has to be stored on Cursor servers.

The legacy rule disallows all storage of data like Memories,… including code and API requests.