Thinking machines are ripe for a world takeover; Tech visionaries are willing to spend to simulate human intelligence, writes Anjana Ahuja

June 10, 2014 6:57 pm

Thinking machines are ripe for a world takeover

By Anjana Ahuja

Tech visionaries are willing to spend to simulate human intelligence, writes Anjana Ahuja

If it looks like a duck and sounds like a duck, then it probably is a duck. That is the inelegant logic behind one of the challenges posed in artificial intelligence: the Turing test, which sets out to answer the question, “can machines think?”

The stroke of genius from Alan Turing, the second world war codebreaker, was to recognise that while actual sentience in machines is virtually impossible to verify, the illusion of sentience is absolutely testable. He proposed that if a machine could “converse” with a person so convincingly that the user thinks they are interacting with a real person, then that machine can be said to think.

According to weekend reports, Turing’s benchmark of artificial intelligence, which dates back to 1960, has been met by a supercomputer disguised as a teenager from Ukraine. In a test devised by the University of Reading, a third of judges having a five-minute text conversation with “Eugene Goostman” believed he was a 13-year-old boy, rather than an advanced natural language computer program. Such advances, the organisers say, will set the scene for a new, sinister kind of cybercrime, in which trusting people are fooled by clever machines into handing over sensitive information.

One may harbour suspicions about this milestone: the competition was organised by a publicity-hungry academic and the result was declared in a press release rather than in a scientific paper; there have been previous claims of success by other groups; the judges may have made overgenerous allowances for a child typing in a second language; why a conversation with any teenager, real or contrived, might be regarded as an appropriate test of sentience.

Scepticism aside, AI is enjoying a resurgence. The goal is to build a machine that thinks like a human, and Google leads a pack of companies keen to see this promise fulfilled. Larry Page, chief executive, championed its recent acquisition of DeepMind, a UK start-up devoted to “deep learning”, in which machines mimic the way a human brain operates. As a concept, deep learning has been around for decades, but only recently have machines come close to the brain’s processing power.

In the past 18 months, Google has also snapped up a dozen robotics companies; the corporation is now the keeper of Cheetah, the world’s fastest legged robot. It has employed the far-out thinker Ray Kurzweil, who predicts the imminent arrival of the “singularity”, the point when artificial intelligence overtakes human intelligence.

In December, Facebook created a new AI laboratory and recruited the renowned computer scientist Yann LeCun to run it. Baidu, the Chinese search engine, has just hired Andrew Ng, Google’s former AI chief, to run its deep learning laboratory in Silicon Valley. Two years ago, Mr Ng taught the Google Brain – a neural network of 16,000 computers – to spot a cat, after training it on thousands of cat images. That feat made headlines because, unlike the rule-based algorithms that dominate computing today, Google Brain had never been fed rules for identifying felines. Google is now pioneering the driverless car, another AI totem.

It feels like an extraordinary moment: after decades of penury, tech visionaries armed with limitless revenues, untethered to risk-averse shareholders and governments, are willing to spend whatever it takes to simulate human intelligence.

That these corporations are coming to dominate the field should not go unnoticed. Forget the cliché of robots taking over; instead, contemplate what happens when bandwidth-limited humans are sidelined in a data-dominated future. We are seeing the inception of an AI-powered autonomous economy: news aggregators are being linked to autonomous financial trading systems, leading to shares being traded between machines. To paraphrase one observer, when machines talk directly to machines, what will we need people for?

Turing anticipated just such a backlash, predicting detractors would argue that a thinking machine could not “be kind, resourceful, beautiful, friendly, have initiative, have a sense of humour, tell right from wrong, make mistakes, fall in love . . . ”

And who knows? Eugene might even agree with him.

The writer was named best science commentator at the 2013 Comment Awards