The announcement by the University of Reading TuringthatEugene Goostman- a program that, as announced over the weekend, passed the famous test "for the first time", which is considered to be a milestone/criterion for when a machine can be considered "thinking".
A particularly "sharp" article by Techdirt states that "Eugene Goostman" by Vladimir Veselov and Eugene Demchenko, which was awarded at an event organized by the university at the Royal Society in London, was not some kind of "supercomputer" but a chatbot, that is, a script specifically designed to imitate human conversation, so it is not "artificial intelligence".
It is also claimed that other such chatbots have been shown to pass the "Turing test" (where a computer "fools" its human interlocutors, making them think they are dealing with a human and not a computer), and that the "success" is due to "misleading" the rules (the fact that it was a 13-year-old boy from Ukraine somehow "covered up" some strange responses).
Also, the publication questions the credibility of one of the organizers, Kevin Warwick of the University of Reading, who is accused of repeated misleading claims.
As reported in a Guardian article, Stéphane Harnand, a professor at the University of Quebec (Montreal), argued that whatever happened at the Royal Society did not constitute "passing" the Turing test. "We have not passed the Turing test. We have not even come close.".
"Turing's thinking was that the best way to explain how the mind works is to design a system that can do everything the mind can do. That includes all of our speech capabilities, as well as our sensory and perceptual capabilities...not for five minutes, but for a lifetime.".
According to a Huffington Post report, many artificial intelligence experts question the validity of the "version" of the Turing test used by the University of Reading.
Gary Marcus, a professor at New York University, said it was far from true intelligence, while Hugh Lebder, the creator of a similar competition, said the time frame for the discussions was too short to reach a sound decision. Douglas Lenat, an artificial intelligence researcher and CEO of Cycorp, commented that “if I teach my parrot to talk a lot, responding to what I say, it may be lovable, but not intelligent.”.
However, John Denning, who worked with Veselov, spoke to the Guardian in support of the program, noting that "I think we passed 'a' Turing test, but I don't know if it was 'the' Turing test. Is Eugene smarter than a person? No. You're not going to put your life in the hands of a 13-year-old who makes jokes and has a weird sense of humor.".
Also, in a statement to PC World, Columbia University professor Tony Jebara hailed the achievement, noting that since its conception, the concept of artificial intelligence has been defined by one milestone after another, "from whether it plays chess to whether it learns to compose sentences.".
"We've said over and over again that 'once a computer can do this, we've created intelligence.' However, every time we reach a milestone, we stand up and say 'this is still not artificial intelligence' and raise the bar even higher. We still have a lot to do and learn about how to make computers intelligent and how our brains work.".
It is noted that the original University of Reading post on the Internet has been amended to describe Eugene as a computer program rather than a “supercomputer.” It also clarifies details about the nature of the competition tests, as conducted at the Royal Society.
