# Role-based access control

GLOSSARY, read 10 Nov 2026

**Definition.** Role-based access control (RBAC) grants permissions to roles, not individuals, so a user can only see and change what their role allows. In a vulnerability platform it limits who can accept risk, edit scoring, or run automation. It also bounds what AI assistants can do on a user's behalf.

Audit evidence usually includes the role list, who holds each role and the change log. Separation of duties, such as the requester not approving their own exception, is the common control test.

Documentation of role limits for AI agents is one of the factors in the AI governance criterion of the index.

## A sourced example

On 2026-11-10 ServiceNow's documentation for AI skills and agentic workflows in Unified Security Exposure Management stated that AI agents use role-based access controls.

## Related terms

- [Multi-tenancy](/glossary/multi-tenancy/)
- [Risk acceptance](/glossary/risk-acceptance/)
- [Vulnerability exception](/glossary/vulnerability-exception/)
- [Plan of action and milestones (POA&M)](/glossary/plan-of-action-and-milestones/)

## Sources

- Source for the example: Role-based access control: https://www.servicenow.com/docs/api/khub/maps/x8~ohSTeWLioGLVI5x3Vgg/topics/RO3F5ODkc6PeIizmnEuGDg/content (read 10 Nov 2026)

Last reviewed 10 Nov 2026. Canonical: https://riskbasedvulnerabilitymanagement.com/glossary/role-based-access-control/
