Artificial intelligence defeats champion of verbal debate for the first time

IBM began developing artificial intelligence for the debate in 2014, as a separate branch of the development of "machine thinking", as opposed to systems with precise calculation. Deep Blue's computer, which beat Garry Kasparov in a game of chess in 1997, operated in a strict field of rules. AlphaGo won out in a go game where winning strategy is blurred in an infinite number of options. And on June 17, 2018, the Project Debater system defeated a person in a verbal dispute!

Artificial intelligence Project Debater was created to realize the historically difficult goal of filtering information sources according to their value, based on the likelihood of misinformation among them, deliberate or accidental. Simply put, he knows how to reason, soberly evaluate arguments and select counter-arguments in order to reveal the information necessary to reach the truth. Only at the same time is he devoid of the influence of emotions, bias, ambiguity and generally does not have his own opinion, as such.

Project Debater's opponent in the demonstration test debate was Noah Obadia, the Israeli National Debate Champion. AI did not have access to the Internet, but it had a knowledge base of hundreds of millions of scientific publications. The topic "Public funding of space exploration" was chosen at random from 100 options, AI was given the command to stand up for, Ovadia opposed. And lost, according to the aggregate audience assessment.

The AI ​​acted prudently and despicably - it agreed with most of the opponent's arguments, but immediately argued that their significance was not so great as to close the question. As a result, Ovadia simply ran out of arguments and time, and the AI ​​showed that it can turn the enemy's weapons against him.