Granting access to systems and data by job role, so each person's permissions come from the role they hold.
Role-Based Access Control (RBAC) decides what each person may open or change by the job they hold in the organization.
Permissions are attached to roles such as "Data Scientist" or "System Administrator", and people are given roles. Changing what a role may do changes it for everyone who holds that role, which keeps access policy manageable as the organization grows.
RBAC carries over to Private AI deployments: a Large Language Model should surface insights only from data the querying user is authorized to view.
In an enterprise RAG system, that means propagating RBAC down to the Vector Database level, so the model cannot retrieve or synthesize documents outside the permissions of the user's role.
See also Private AI, Large Language Model, Data Governance, Vector Database, Air-Gapped Environment and Zero Trust Architecture.