Overview
Agent identity is how IBM Verify recognizes, authenticates, and governs AI agents operating in your environment. As organizations adopt AI agents to automate tasks and interact with services, managing these non-human identities becomes a security requirement — not just an operational one.
IBM Verify provides the tools to onboard, manage, and secure AI agents so they operate with the right level of access and remain accountable members of your digital workforce.
What is an AI agent?
An AI agent is an autonomous software program that perceives its environment, makes decisions, and takes actions to achieve a specific goal. Unlike traditional applications that follow fixed logic, AI agents can plan, reason, and act dynamically — sometimes on behalf of a user, sometimes independently.
Why agents need dedicated identities
AI agents are not users. Assigning a human user account to an agent creates security gaps:
- Human identity policies do not apply correctly to agents
- Access becomes difficult to audit and govern
- Agents may accumulate broader access than they need
IBM Verify assigns agents their own identity constructs, enabling least-privilege access, clear accountability, and lifecycle governance.
What you can do
With agent identity in IBM Verify, you can:
- Onboard agents — Register AI agents before they can authenticate or access resources
- Issue identity tokens — Use the JWT Bearer or client credentials grant flow to authenticate agents
- Enable on-behalf-of (OBO) access — Allow agents to act on behalf of a user with a scoped token
- Issue privileged tokens — Grant agents elevated access when acting on behalf of a user
- Connect agents to tools — Integrate agents with direct tools, MCP servers, or IBM Secrets Vault
- Audit and comply at the agent level — Track agent activity and enforce compliance policies against individual agent identities, giving you a clear record of what each agent accessed, when, and on whose behalf