To train its AI agent to recognize and respond to a new type of customer request, what should a company configure?

The correct approach for training an AI agent to recognize and respond to a new type of customer request involves configuring Custom Topics and Custom Agent Actions. Custom Topics allow the AI to identify and categorize different customer requests, making it easier for the system to understand the intent behind the queries it receives. By defining these specific topics, the AI can be finely tuned to recognize various nuances in customer interactions.

Additionally, Custom Agent Actions help the AI determine how to appropriately respond once a request has been classified under a certain topic. This combination ensures that the AI is not just passively responding but is actively engaging based on the context of the inquiry.

In contrast, allowing AI to generate responses freely could lead to inconsistencies and responses that don’t align well with the company’s objectives or customer expectations. Relying solely on predefined chatbot scripts limits the AI’s ability to learn and adapt, which is essential for handling new and varied requests effectively. Disabling AI interactions for new customer inquiries completely undermines the purpose of leveraging AI to improve customer service and support efficiency. Thus, configuring Custom Topics and Custom Agent Actions is essential for effective AI training.