Flow authors can now enable virtual agents to provide AI-generated answers based on the article that best matches the customer query. Answer generation combines the capabilities of a language model (LLM) with knowledge bases to retrieve relevant data from a single knowledge article and then use that data to generate contextual, accurate, and dynamic responses.

Answer generation improves digital bot responses in dynamic and data-heavy environments and is most beneficial for content that changes dynamically or large amounts of data. This feature ensures that customers receive direct, accurate responses without needing to read through lengthy articles to find the information they need.

To access answer generation, follow these steps:

1. Click Menu > Orchestration > Architect.
2. Open an existing bot flow or create a new one.
3. Confirm that the flow is upgraded to Virtual Agent.
4. Under Natural Language Understanding, click Knowledge.
5. Select a knowledge base.
6. Click the Answer tab.
7. Under Single Answer Introduction, select the wording that the bot uses before returning a single knowledge answer. 
8. Under Content Type, to create an AI-generated answer based on the article that best matches the customer query, click Generative.
9. Under Full Article Label, add the label that the user clicks to view the full article and not just the highlighted answer. Conversely, you can perform these steps to build an expression for the label:
   1. Click Expression Mode .
   2. To build an answer, select one of these options:
      • Click Expression .
      • Click Communication Sequence Builder .
      • Click Rich Text Builder .
10. Under Flow Progression Label, add the label that the user clicks to continue to progress through the bot flow. Conversely, you can perform these steps to build an expression for the label:
    1. Click Expression Mode .
    2. To build an answer, select one of these options:
       • Click Expression .
       • Click Communication Sequence Builder .
       • Click Rich Text Builder .
    3. Click Save.

For more information, see Add knowledge to your bot flow.