Overview of IBM Content Assistant

The Content Assistant provides natural language insights about business content in your IBM® FileNet® Content Manager repository. It saves time by avoiding manual, repetitive review of document content and allows you to quickly find information across large volumes of content.

IBM Content Assistant includes generative AI capabilities that can summarize document content and provide accurate answers to questions within the context of the selected business content in a IBM FileNet Content Manager repository.

These capabilities are based on the IBM watsonx Large Language Model (LLM) that is coupled with context-specific Retrieval-Augmented Generation (RAG). The RAG augments the LLM with context-specific knowledge base and information that is hosted in the IBM FileNet Content Manager repository.

The following diagram illustrates how the IBM Content Assistant works:

A user submits a query to the IBM Content Assistant and a query object is created in the Content Platform Engine (CPE) object store. The CPE server calls the generative AI connector service to find business documents that match the user query. It then extracts relevant chunks of documents that are related to the query and submits the user question to the IBM watsonx LLM. The LLM provides the CPE API caller with the response that is updated in a query object. The CPE API caller can include IBM Content Navigator and other CPE applications.