Glossary

OIDC (OpenID Connect)

The protocol behind “Sign in with Google” and most modern single sign-on. OIDC adds an identity layer to OAuth 2.0 so an app learns who the user is, not just what it may access on their behalf.

General definition

OIDC (OpenID Connect) is a thin identity layer standardised by the OpenID Foundation in 2014 on top of the OAuth 2.0 authorisation framework. OAuth 2.0 on its own answers “may this app access this resource?” and says nothing reliable about who the user is. OIDC fixes that by defining an ID token, a signed JWT issued by the identity provider (the OpenID Provider) to the application (the Relying Party) with claims such as iss (issuer), sub (a stable user identifier), aud (the intended app), exp and nonce, plus optional profile claims like name and email.

  • Authorization code flow with PKCE: the recommended flow for web, mobile and single-page apps; the browser is redirected to the provider, the app exchanges a one-time code for tokens over a back channel
  • Discovery: the provider publishes its endpoints and signing keys at /.well-known/openid-configuration, so clients configure themselves
  • UserInfo endpoint: returns additional claims about the authenticated user when the app presents an access token
  • Scopes: openid switches OIDC on; profile, email and others request particular claims
  • Logout: optional specifications cover front-channel, back-channel and RP-initiated logout

The usual comparison is with SAML, the older XML-based enterprise standard. Both deliver single sign-on from a central identity provider. SAML is browser-centred, verbose and deeply embedded in large enterprise IdPs; OIDC uses JSON and REST, fits mobile apps and APIs naturally, and is what consumer providers (Google, Apple, Microsoft) and newer workforce IdPs lead with. Most enterprise IdPs support both, and buyers in regulated sectors increasingly accept either as long as the application delegates authentication to the corporate directory and honours deprovisioning.

Two things OIDC does not do. It does not by itself provision or remove user accounts in the application; that is left to just-in-time creation at first login or to a provisioning protocol such as SCIM. And it does not define what the user may do inside the application; roles and permissions remain the application’s job, informed by claims or group memberships the provider chooses to share.

In the Ethora ecosystem

Ethora’s social sign-on options (Google, Apple, Facebook and others) are OAuth and OIDC integrations: the user authenticates with the provider, Ethora receives the identity claims and issues its own short-lived JWT chat tokens for the session. On dedicated deployments the same delegation pattern can extend to enterprise SSO against the customer’s identity provider, so staff use their corporate login and access is revoked centrally when they leave.

Many customers avoid a visible Ethora login altogether. With API-provisioned end users, the customer’s own backend, which already authenticated the person through OIDC, SAML or anything else, creates the chat user and obtains a chat token for them, so the messaging inside a patient portal or banking app inherits whatever identity system the host application trusts. Roles inside Ethora are then set through the admin panel and API rather than inferred from the provider.

Get started

Chat and AI built for regulated industries

Ethora gives you audit trails, retention controls, a Trust & Safety layer and dedicated deployments in your own cloud. Talk to our team.

Start Free
Free tier available Enterprise SLA No vendor lock-in