How we use your conversations, what controls you have, and what's different on Organization plans.
11 мая 2026
It depends on your plan and the Improve the model for everyone setting. Here's exactly how it works.
By default, your conversations may be used for model training, product improvement, and anonymised research within Magisterium AI. You can opt out at any time:
/account/data-controls).
When this setting is off, new chats are flagged so they aren't used for model improvement.
Chats created under an Organization or Large Organization plan are automatically excluded from training and product improvement — the toggle is fixed and can't be turned on. Members of an organization who also have personal chats keep those personal chats under their own setting.
If you share a chat via the Share menu, it becomes accessible to anyone with the URL. Shared chats may be seen by others if the URL is posted or forwarded. Only share what you're comfortable being public.
When you click thumbs-up or thumbs-down on an answer (see Giving feedback on a response), your rating and any comment are reviewed by our team to evaluate and improve the product.
Personalized memory stores short facts derived from your chats (e.g. "User is a catechist"). These are stored for you and used to tailor your future answers. They live with the rest of your account data and are subject to the same training-consent toggle. See Personalized Memory.
For full legal detail, see our Privacy Policy.
Пошук дапамогі