I am having trouble defining specific items to be out of scope for an AVA. It does a good job of not allowing customer questions that would be outside of it's domain, but it tries to be overly helpful.
For example, I have specified in the Role and in the Guidelines that it should not attempt to troubleshoot an issue or to suggest fixes. I am having a really hard time having it follow these instructions. Any thoughts on wording, formatting, etc that would help it not offer to do something (on its own) that I don't want it to?
#ConversationalAI(Bots,VirtualAgent,etc.)------------------------------
Kevin Goodwin
Senior Consultant
------------------------------