> ## Documentation Index
> Fetch the complete documentation index at: https://docs-staging-ai-docs-migration.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Identity for B2B

> Secure your AI agents with user authentication, delegated authorization, human-in-the-loop controls with CIBA, and fine-grained authorization in RAG pipelines.

<Card title="Secure your AI Agents with Auth0. Authenticate users, call APIs on their behalf with OAuth 2.0, implement human-in-the-loop controls with CIBA, and enforce document-level access control in RAG pipelines." icon="sparkles" iconType="solid" horizontal />

<Frame>
  <img src="https://mintcdn.com/docs-staging-ai-docs-migration/KLtKgmlAEqAze78X/docs/images/ai/Hero-diagram.png?fit=max&auto=format&n=KLtKgmlAEqAze78X&q=85&s=78c3dd86a36c3f3f33e21b00534f708c" alt="AI Identity for B2B diagram" width="2600" height="1135" data-path="docs/images/ai/Hero-diagram.png" />
</Frame>

### User authentication

Integrate Auth0's [Universal Login](/docs/authenticate/login/auth0-universal-login) into your AI agents to verify user identity, with support for social, enterprise, and custom identity providers. Leverage standards like OAuth 2.0 and OpenID Connect to secure both interactive and headless AI agents. [Learn More](./intro/user-authentication).

### Call first-party APIs on a user's behalf

Enable AI agents to securely call first-party APIs on the user's behalf using OAuth 2.0. This ensures that the user's context is maintained, preventing agents from having broad, unrestricted access to user data. [Learn More](./intro/call-first-party-apis-on-users-behalf).

### Call third-party APIs on a user's behalf

Securely connect AI agents to external tools and services like Google, Slack, or GitHub. Auth0's Token Vault manages obtaining, storing, and refreshing API tokens, allowing your agents to interact with third-party APIs without handling sensitive credentials or API keys directly. [Learn More](./intro/call-third-party-apis-on-users-behalf).

### Asynchronous authorization

Implement "human-in-the-loop" workflows for actions that are long-running or require user consent. Using standards like [CIBA (Client-Initiated Backchannel Authentication)](/docs/get-started/authentication-and-authorization-flow/client-initiated-backchannel-authentication-flow), agents can request approval for sensitive operations via push notifications, SMS, or email, even when the user is not actively using the application. [Learn More](./intro/asynchronous-authorization).

### Authorization for RAG

Enforce fine-grained, document-level access control within your Retrieval Augmented Generation (RAG) pipelines using [Auth0 FGA](https://auth0.com/fine-grained-authorization). This ensures that when a Large Language Model (LLM) generates a response, it only uses data that the current user is authorized to view, preventing [sensitive data leakage](https://genai.owasp.org/llmrisk/llm022025-sensitive-information-disclosure/) and ensuring compliance. [Learn More](./intro/authorization-for-rag).

## Get started

Follow our [quickstart guides](/ai/docs/get-started/overview) for a step-by-step tutorial on implementing authentication and authorization in your AI agents.

## Sample applications

Explore our [sample applications](/ai/docs/sample-apps/sample-apps) and templates.

## Frameworks & SDKs

Explore our [SDKs](/ai/docs/sdks/overview) for the different frameworks supported by AI Identity for B2B.