On Automating Conversations

TL;DR

This paper presents Chorus, a crowd-powered conversational system that was gradually automated using the Evorus framework, demonstrating real-world deployment and addressing automation challenges in human-AI conversations.

Contribution

It introduces a top-down approach to automate a crowd-powered conversational assistant over time, with real-world deployment insights.

Findings

Over 420 users engaged with Chorus

More than 2,200 conversation sessions conducted

Demonstrated successful automation of a conversational agent

Abstract

From 2016 to 2018, we developed and deployed Chorus, a system that blends real-time human computation with artificial intelligence (AI) and has real-world, open conversations with users. We took a top-down approach that started with a working crowd-powered system, Chorus, and then created a framework, Evorus, that enables Chorus to automate itself over time. Over our two-year deployment, more than 420 users talked with Chorus, having over 2,200 conversation sessions. This line of work demonstrated how a crowd-powered conversational assistant can be automated over time, and more importantly, how such a system can be deployed to talk with real users to help them with their everyday tasks. This position paper discusses two sets of challenges that we explored during the development and deployment of Chorus and Evorus: the challenges that come from being an "agent" and those that arise from the subset of conversations that are more difficult to automate.