IBM's AI discussion system is able to compete with expert debaters


Photo: Public Domain CC0

IBM has developed a system based on artificial intelligence, designed to participate in debates with people. In their paper published in the journal Nature, the team members describe their system and how well it performs when pitted against human opponents. Chris Reed, from the University of Dundee, published an article, "News and Views," in the same journal issue, which describes the history and development of artificial intelligence technology, based on the types of logic used in human arguments, and a new system developed by IBM.


As Reed notes, the discussion is a skill people have been honing for thousands of years. This is generally considered a type of discussion in which one or more people try to convince others that their opinion on a topic is correct. In the course of this new effort, the IBM team created an artificial intelligence system designed to discuss with people in a live environment. He listens to moderators and opponents and responds with a female voice.


In most debates, people presenting an argument tend to refer to others who can support their claims. They may note previous research, or quote well-known phrases used by people respected in the field of argumentation. IBM's system, known simply as Project Debater, scans the internet for such arguments and uses them in a way that it finds compelling.



Most debates also usually involve participants trying to knock down their opponent's arguments. To perform such tasks, Project Debater uses Watson, an IBM system that beats the participants of the game show "Jeopardy" to listen to the arguments of opponents, and then look for refutations that have been given by other similar claims.


IBM began testing the system back in 2019 when it participated in a debate with Harish Natarajan, an expert debater. The audience agreed that Project Debater did not beat Natarajan, but the same audience also agreed that he did very well. In a later test, Project Debater was asked to convince a group of viewers that telemedicine was a good idea. Most of the panelists found that the AI system did indeed change its position on the topic – a possible indication that AI systems may one day soon play a role in human debates similar to those taking place on social media sites.

4 views0 comments

Ads

Ads

FYBERUS

  • Facebook
  • Twitter
  • Instagram

Copyright © 2020 Fyberus WebSite. A Fyberus. All rights reserved. Reproduction in whole or in part without permission is prohibited.