Currently, as I’ve understood it, the AI conversation ends as soon as a handoff is triggered. This is problematic in at least 2 scenarios:
Scenario 1: No live agent available
It doesn’t seem to matter if there is an agent available or not. The chat seems to end, leaving the user without any assistance and forcing them to repeat the chat if they want to give the AI Agent another chance.
Scenario 2: The "stuff" is hitting the fan.
The user demands emergency contact with a live agent, which immediately triggers a handoff for our AI agent. While they wait, they are not being assisted.
While an AI agent resolving an emergency is unlikely, it might still surface information that helps the live agent narrow things down faster and make the user feel cared for while being essentially on hold. Even just a bit of light fishing for info, while the user waits, could be helpful.