The creator of chatbot Mitsuku, Steve Worswick, won the 2018 Loebner prize final and the $4000 prize money for the fourth time on September 8, making him a joint all time champion, along with x and y. The Loebner prize is based on a test by the computer scientist and World War Two code breaker, Alan Turing, who is widely considered to be the father of Artificial Intelligence (AI). The test was designed by the late Turing to see if machines can think, however the Loebner prize interpretation has been both highly influential and criticized.
In the competition, judges simultaneously have anonymous text-based conversations with a chatbot and a human being via a computer and have have to guess which one is which.
None of the chatbots competing in the 2018 finals at Bletchley Park, UK, managed to fool a judge into believing it was human so they were ranked on how ‘human-like’ they were. Out of a possible score of 100 percent, Mitsuku came first place with 33, Tutor second with 30, Colombina third with 25 and Uberbot fourth with 23 (talk to the bots via the links).
How relevant is the Loebner prize within the AI field?
It has been called “a useful way to chart the progress of AI” by AI expert and former judge Noel Sharkey, and an “obnoxious and unproductive annual publicity campaign” by the late leading AI and philosophy researcher, Marvin Minsky.
The Loebner prize was first launched in 1990 by the New York businessman Hugh Loebner, who died in 2016.