What should you add to your Power Virtual Agents to prevent escalations to a human agent when your chatbot cannot determine the user's intent?

A fallback topic in Power Virtual Agents is specifically designed to activate when a user's input does not match any existing defined topics. This allows the bot to provide a structured response, ask for clarification, or redirect the user appropriately instead of failing or escalating.

System topics are pre-built conversational elements like greetings or farewells, but they do not handle scenarios where the user's overall intent is ambiguous or unknown.

Variables are used for storing and manipulating data during a conversation but do not by themselves determine how the bot responds to unmatchable user input.

Entities are used to identify and extract specific types of information (e.g., dates, cities) from user utterances, not to manage situations where the general intent is unclear.