A context graph is a connected model of an enterprise's information for AI use — mapping entities, relationships, sensitivity, provenance, policy and lineage — so that governance decisions can be made about meaning and connection rather than about individual files.
What is a context graph?
Documents, records, people, systems and agents are related in ways a file system cannot express, and those relationships are what make aggregation risk possible. A context graph makes the relationships explicit and attaches sensitivity, ownership, policy and lineage to them, giving the policy engine something to reason over. It is also what makes a governance decision explainable after the fact: the graph shows what was connected to what, and why the system concluded that the connection mattered.