The purpose of these classifications is to abstract the underlying identity providers (IdPs) from their presumed level of trust. The outcome is that you will be able to build policies that target levels of trust instead of specific IdPs.

Defining trust levels has the following distinct parts: