Openstream: Taking a Knowledge-centric, Collaborative Dialogue Approach to Conversational AI

An Intellyx Brain Candy Brief

It seems that conversational AI agents are the new must-have tech industry accessory. Not only does every major platform player now have one, but there are dozens (if not hundreds) of companies delivering some form of a chatbot or intelligent AI agent to the market. And most of them function in precisely the same way: focusing on the conversational bit.

While this conversational focus would appear to make sense on the surface, Openstream believes that the path to a truly effective intelligent agent doesn’t start with the conversation (or what it calls the interaction) but, instead, with knowledge. The company’s platform employs an automated knowledge and entity extraction process that ingests an organization’s documents, emails, transcripts, audio, video, etc., to create a knowledge graph. It then applies domain-specific models to generate an ontology unique to the organization.

To this knowledge layer, it adds context and interaction management to create its intelligent agent, which it calls Eva. The company believes that this approach enables it to achieve its goal of creating what it calls a “collaborative dialogue” with users in which it seeks to understand their goal, assess if it is reasonable and achievable, and then either assist them if it is or provide alternatives if it isn’t. It represents a fresh approach that the company believes will deliver superior results in its core industry sectors of insurance, financial services, healthcare, and manufacturing and aerospace.

©2021 Intellyx LLC. Intellyx advises companies on their digital transformation initiatives, and helps vendors communicate their agility stories. At the time of this writing, none of the companies mentioned are an Intellyx customer. To be considered for a Brain Candy article, email us at pr@intellyx.com.

SHARE THIS: