Privacy label

A privacy label is a short tag attached to a source or record that tells the assistant what it may do with that material, with a cautious local-only default so sensitive content stays put until you approve wider use. You attach one tag per item, such as "open" for things you would happily share, "internal" for material that can flow between your own tools, and "private" for items the assistant reads but never copies into an outgoing draft. Say you keep a notes folder where most files are tagged "internal" but one file holding account numbers is tagged "private." When you ask the assistant to draft a status update, it pulls freely from the internal notes and leaves the private file out of the text it produces. The label travels with the record, so the rule applies every time that record is touched.

Builder example

Without a privacy label, an assistant treats every readable file as fair game for whatever output you request, so one careless summary can paste a home address or a password into a shared document. Tagging your sources fixes the default: tell the assistant to honor the label on each record, keep "private" material out of anything it sends or publishes, and ask before promoting an item to a more permissive tag. A weekly review of what is labeled what keeps the boundary current as new records arrive.

Common confusion: A privacy label controls what the assistant may do with a record, while a permission scope controls which account the assistant may reach into at all. Scope decides whether it can open your email; the label on a given message decides whether its contents may travel into an outgoing draft.