8th June 2014

Turing Test passed? Researchers claim breakthrough in artificial intelligence

Researchers are claiming a major breakthrough in artificial intelligence with a machine program that can pass the famous Turing Test.

At the Royal Society in London yesterday, an event called Turing Test 2014 was organised by the University of Reading. This involved a chat program known as Eugene being presented to a panel of judges and trying to convince them it was human. These judges included the actor Robert Llewellyn – who played robot Kryten in sci-fi comedy TV series Red Dwarf – and Lord Sharkey, who led a successful campaign for Alan Turing's posthumous pardon last year. During this competition, which saw five computers taking part, Eugene fooled 33% of human observers into thinking it was a real person as it claimed to be a 13-year-old boy from Odessa in Ukraine.

In 1950, British mathematician and computer scientist Alan Turing published his seminal paper, "Computing Machinery and Intelligence", in which he proposed the now-famous test for artificial intelligence. Turing predicted that by the year 2000, machines with 10 GB of storage would be able to fool 30% of human judges in a five-minute test, and that people would no longer consider the phrase "thinking machine" contradictory.

In the years since 1950, the test has proven both highly influential and widely criticised. A number of breakthroughs have emerged in recent times from groups claiming to have satisfied the criteria for "artificial intelligence". We have seen Cleverbot, for example, and IBM's Watson, as well as gaming bots and the CAPTCHA-solving Vicarious. It is therefore easy to be sceptical about whether Eugene represents something genuinely new and revolutionary.

Professor Kevin Warwick (who also happens to be the world's first cyborg), comments in a press release from the university: "Some will claim that the Test has already been passed. The words 'Turing Test' have been applied to similar competitions around the world. However, this event involved the most simultaneous comparison tests than ever before, was independently verified and, crucially, the conversations were unrestricted. A true Turing Test does not set the questions or topics prior to the conversations. We are therefore proud to declare that Alan Turing's Test was passed for the first time on Saturday."

Eugene's creator and part of the development team, Vladimir Veselov, said as follows: "Eugene was 'born' in 2001. Our main idea was that he can claim that he knows anything, but his age also makes it perfectly reasonable that he doesn't know everything. We spent a lot of time developing a character with a believable personality. This year, we improved the 'dialog controller' which makes the conversation far more human-like when compared to programs that just answer questions. Going forward, we plan to make Eugene smarter and continue working on improving what we refer to as 'conversation logic'."

Is the Turing Test a reliable indicator of intelligence? Who gets to decide the figure of 30% and what is the significance of this number? Surely imitation and pre-programmed replies cannot qualify as "understanding"? These questions and many others will be asked in the coming days, just as they have been asked following similar breakthroughs in the past. To gain a proper understanding of intelligence, we will need to reverse engineer the brain – something which is very much achievable in the next decade, based on current trends.

Regardless of whether Eugene is a bona fide AI, computing power will continue to grow exponentially in the coming years, with major implications for society in general. Benefits may include a 50% reduction in healthcare costs, as software programs are used for big data management to understand and predict the outcomes of treatment. Call centre staff, already competing with virtual employees today, could be almost fully automated in the 2030s, with zero waiting times for callers trying to seek help. Self-driving cars and other forms of AI could radically reshape our way of life.

Downsides to AI may include a dramatic rise in unemployment as humans are increasingly replaced by machines. Another big area of concern is security, as Professor Warwick explains: "Having a computer that can trick a human into thinking that someone – or even something – is a person we trust is a wake-up call to cybercrime. The Turing Test is a vital tool for combatting that threat. It is important to understand more fully how online, real-time communication of this type can influence an individual human in such a way that they are fooled into believing something is true... when in fact it is not."

Further into the future, AI will gain increasingly mobile capabilities, able to learn and become aware of the physical world. No longer restricted to the realms of software and cyberspace, it will occupy hardware that includes machines literally indistinguishable from real people. By then, science fiction will have become reality and our civilisation will enter a profound, world-changing epoch that some have called a technological singularity. If Ray Kurzweil's ultimate prediction is to be believed, our galaxy and perhaps the entire universe may become saturated with intelligence, as formally lifeless rocks are converted into sentient matter.

Comments »