A Turing Test – the biggest ever staged, according to New Scientist – took place on 23rd June at Bletchley Park to mark the Turing centenary.
The test involved 150 conversations, 30 judges, 25 humans and five chatbots. The article points out that the Loebner Prize typically involves four judges and four chatbots. The contest was won by ‘Eugene Goostman‘, “a chatbot with the personality of a 13-year-old boy” which fooled judges 29% of the time.
Eugene, which has won second prize in the Loebner Prize several times, was created by Vladimir Veselov to have a character of “a 13-year-old boy living in Odessa, Ukraine. He has a pet guinea pig and a father who is a gynaecologist”. Why thirteen? Veselov is quoted saying “thirteen years old is not too old to know everything and not too young to know nothing”. And why a specific character? AI researcher John Barnden is quoted saying “I think any appearance of a particular personality is likely to have a persuasive effect on judges” – though the article stresses that he cautions against concluding that this was Eugene’s edge: “for that you would have to compare two versions of the same bot, but in one case with personality suppressed”.