The media update team
looks at what makes a Turing Test unique and how it will come under increasing strain, due to AI, in the years to come.
The Turing Test evaluates a computer being able to talk like a person
The Turing Test was created to determine whether or not a computer could be classed as intelligent. It tests a machine’s ability to behave in a way that is inseparable from humans.
It works by having human judges sit on one side of a computer screen. The judges have to chat to several “people” on the other side of the screen. Most of the people speaking to the judge will be human.
However, one of those “people” will actually be a chatbot. The chatbot’s objective is to convince the human judges that they are speaking to a real person. If it does this, it has passed the Turing Test.
The Turing Test has evolved along with AI
The Turing Test has evolved since it was first theorised by Turing himself. Newer versions of the Turing Test can be described as “a Turing Test” or a "Turing-style test".
The foundations of the newer Turing-style tests are based on the original, but they assess the intelligence of the machine much more stringently. This is very important because AI will continue to develop. The original Turing Test, created over 60 years ago, might lose relevance as AI evolves
In June 2014, the BBC reported
that a machine succeeded in passing the Turing Test. The machine, called Eugene, was a chatbot that managed to convince a third of the judges it spoke to that it was a human being. Eugene was given the persona of a 13-year-old boy.
The Loebner Prize
, launched by inventor Hugh Loebner in 1990, requires a human and computer to have a 25-minute conversation with each other in four different languages. The machine will pass the human test if it tricks half of the judges into thinking it is a person.
An even more difficult test
was created by Lotus founder Mitch Kapor[AM2] .
It arose from a USD $20,000 bet Kapor made with futurologist Ray Kurzwil
, which is still unsettled today. In the test, the robot and three human judges must have a two-hour conversation with each other. The robot must convince two of the three judges that it is human and that it is more human than at least two of the judges. Kurzwil
is increasingly confident
he will win the bet. AI keeps evolving
, so maybe Kurzwil’s
confidence is well placed.
Want to stay up to date with the latest media news? Subscribe to our newsletter
AI plays an important role in placing online adverts next to content. Is your brand in the right place? Read more in our article, What is brand safety?
Image courtesy of Elliot Brown, under this licence.