Answer from Documents helps answer end-user questions from unstructured PDF documents without the need to extract individual FAQs and train them. This feature leverages LLM and Generative AI models from OpenAI to process uploaded documents and user queries and generate answers from the documents. Simply upload the documents covering topics you anticipate from end-users. The content from the documents and user utterances (after applying PII redactions) are shared with OpenAI for curating the answer. This is a beta feature and is available only for the English language.

Go to Build > Knowledge AI > Answer from Documents and click Get Started. Follow onscreen instructions to enable the feature and upload documents.

Notes:

You can upload a maximum of 10 PDF documents, and each document should not be more than 5 MB.
Answer from documents acts as a fallback when none of the three NLP engines are able to identify any intent.