That argument is a non-starter. Surely somebody will build and market such solutions once TLS 1.3 becomes widely used.

The protocol must be finalized before enterprise-grade products can use it, not the other way around.