Learn · Identity
What is agentic identity? Agent identity, defined
Agentic identity is the identity an AI agent holds for itself, as opposed to borrowing a person's login or a shared API key. A usable agent identity has three parts: it is bound to the machine the agent runs on, it names the person who answers for the agent, and it is checked every time the agent acts.
What the term covers
People use several names for the same need: agentic identity, agent identity, AI agent identity, autonomous agent identity, digital actor identity and first-class agent identity. Agentic identity management (AIM) is the practice of enrolling, approving, checking and revoking those identities. They all point at one gap: an agent that acts on its own is neither a person nor a classic service account.
A first-class agent identity is one that is not borrowed. The agent has its own key, its own record and its own audit trail, so you can tell which agent did what, and stop one without breaking a person's account.
The three parts of a usable agent identity
- Bound to its machine. The agent's own key is created and kept on the host it runs on. Copy the agent to another machine and you get the same code with a different fingerprint, and no token.
- Accountable. A named person, the principal, signs off on the agent before it acts. Every action traces to the agent and to that person.
- Checked every time. The agent proves who it is on each token request and signs a single-use proof for each call, so a copied token is useless on its own.
What it is not
It is not a login for a human, a shared service account or a static API key. Those identify whoever holds the secret, not the agent. It is also not a judgement about what the model says: identity answers which agent, for whom, and may it, not whether an answer is right.
How Oathera helps
Oathera gives every AI agent an identity bound to the machine it runs on, with a named person who answers for it, and checked every time it acts. Oathera Agent Enrollment creates the identity, Oathera Agent Enforcement checks each call, and the Oathera Identity Service issues identity tokens and records every decision. The agent's identity is the key digest of its anchor key.
Oathera does not verify people, judge model output, inspect or rewrite prompts, or replace your identity provider. Your identity provider and proofing sources do the first; Oathera is the layer under them for agents.
FAQ
What is agentic identity?
The identity an AI agent holds in its own right: bound to the machine it runs on, with a named person who answers for it, and checked every time it acts. It replaces borrowed keys, shared service accounts and someone else's login.
Is agent identity the same as non-human identity (NHI)?
Agent identity is a kind of non-human identity. Classic NHI covers service accounts, API keys and certificates. An autonomous agent adds two needs those do not cover: a named accountable person and a check on every action.
What is Know Your Agent (KYA)?
Know Your Agent is the category: verifying which agent is acting, who answers for it and whether it may act. Oathera is the company and product that does this.
Does Oathera verify who the person is?
No. Oathera relies on your identity provider and proofing sources to verify people. It records which verified person approved each agent.
Related: What is Know Your Agent (KYA)? · Per-agent identity vs shared API keys · What agentic identity governance looks like
Become a design partner Watch the simulator More guides
← Back to Learn