TY - JOUR

T1 - An autonomous debating system

AU - Slonim, Noam

AU - Bilu, Yonatan

AU - Alzate, Carlos

AU - Bar-Haim, Roy

AU - Bogin, Ben

AU - Bonin, Francesca

AU - Choshen, Leshem

AU - Cohen-Karlik, Edo

AU - Dankin, Lena

AU - Edelstein, Lilach

AU - Ein-Dor, Liat

AU - Friedman-Melamed, Roni

AU - Gavron, Assaf

AU - Gera, Ariel

AU - Gleize, Martin

AU - Gretz, Shai

AU - Gutfreund, Dan

AU - Halfon, Alon

AU - Hershcovich, Daniel

AU - Hoory, Ron

AU - Hou, Yufang

AU - Hummel, Shay

AU - Jacovi, Michal

AU - Jochim, Charles

AU - Kantor, Yoav

AU - Katz, Yoav

AU - Konopnicki, David

AU - Kons, Zvi

AU - Kotlerman, Lili

AU - Krieger, Dalia

AU - Lahav, Dan

AU - Lavee, Tamar

AU - Levy, Ran

AU - Liberman, Naftali

AU - Mass, Yosi

AU - Menczel, Amir

AU - Mirkin, Shachar

AU - Moshkowich, Guy

AU - Ofek-Koifman, Shila

AU - Orbach, Matan

AU - Rabinovich, Ella

AU - Rinott, Ruty

AU - Shechtman, Slava

AU - Sheinwald, Dafna

AU - Shnarch, Eyal

AU - Shnayderman, Ilya

AU - Soffer, Aya

AU - Spector, Artem

AU - Sznajder, Benjamin

AU - Toledo, Assaf

AU - IBM Research AI

PY - 2021

Y1 - 2021

N2 - Artificial intelligence (AI) is defined as the ability of machines to perform tasks that are usually associated with intelligent beings. Argument and debate are fundamental capabilities of human intelligence, essential for a wide range of human activities, and common to all human societies. The development of computational argumentation technologies is therefore an important emerging discipline in AI research1. Here we present Project Debater, an autonomous debating system that can engage in a competitive debate with humans. We provide a complete description of the system’s architecture, a thorough and systematic evaluation of its operation across a wide range of debate topics, and a detailed account of the system’s performance in its public debut against three expert human debaters. We also highlight the fundamental differences between debating with humans as opposed to challenging humans in game competitions, the latter being the focus of classical ‘grand challenges’ pursued by the AI research community over the past few decades. We suggest that such challenges lie in the ‘comfort zone’ of AI, whereas debating with humans lies in a different territory, in which humans still prevail, and for which novel paradigms are required to make substantial progress.

AB - Artificial intelligence (AI) is defined as the ability of machines to perform tasks that are usually associated with intelligent beings. Argument and debate are fundamental capabilities of human intelligence, essential for a wide range of human activities, and common to all human societies. The development of computational argumentation technologies is therefore an important emerging discipline in AI research1. Here we present Project Debater, an autonomous debating system that can engage in a competitive debate with humans. We provide a complete description of the system’s architecture, a thorough and systematic evaluation of its operation across a wide range of debate topics, and a detailed account of the system’s performance in its public debut against three expert human debaters. We also highlight the fundamental differences between debating with humans as opposed to challenging humans in game competitions, the latter being the focus of classical ‘grand challenges’ pursued by the AI research community over the past few decades. We suggest that such challenges lie in the ‘comfort zone’ of AI, whereas debating with humans lies in a different territory, in which humans still prevail, and for which novel paradigms are required to make substantial progress.

UR - http://www.scopus.com/inward/record.url?scp=85102734165&partnerID=8YFLogxK

U2 - 10.1038/s41586-021-03215-w

DO - 10.1038/s41586-021-03215-w

M3 - Journal article

C2 - 33731946

AN - SCOPUS:85102734165

VL - 591

SP - 379

EP - 384

JO - Nature

JF - Nature

SN - 0028-0836

IS - 7850

ER -