Competition claims a computer has passed the Turing test
DOI: 10.1063/PT.5.027996
Techdirt : During a competition staged at the University of Reading in the UK on 7 June, a panel of judges determined that a software program called Eugene Goostman passed the iconic Turing test for artificial intelligence. The competition involved 30 judges who conducted keyboard-only interviews with both humans and machines for five-minute intervals. If a computer was mistaken for a human more than 30% of the time, it would pass the test. The competition has been criticized for its many flaws, including the fact that the winner is not a supercomputer but rather a software program that simulates a 13-year-old Ukrainian boy as a way of masking any odd responses. Critics also point out that the ability to speak is not necessarily a measure of intelligence; the more important criterion is cognitive ability.
