23 September 2009 | Team Tamar

The rise of the machine

As computer applications become more logic-oriented and increasingly intelligent, one must begin to wonder – will the day come when artificial intelligence will outstrip human intelligence?

Computer-based systems are continuously becoming more intelligent in all areas. From web-applications, automated control systems in cars, to military and medical applications and even space odysseys, the machines speed and logic-oriented control has proved most useful to us mere mortals.

It’s not news that computers have been able to simulate intelligence in areas such as chess-playing –where they use pure and simple calculation of all the possible options and their lightning quick calculation speed to their advantage.
In fact, in the kind of jobs that require this sort of massive calculation proficiency, computers have been proven irreplaceable. However, computers are still behind in many aspects of human intelligence, such as emotional intelligence and creative thinking… but they are catching up. Fast.

The Turing Test is a test which analyzes the ‘humanistic’ capabilities of a computer. The Loebner Prize is a competition which uses the Turing Test to gauge a machine’s intelligence, and in October 2008 the bronze medal (the highest award given to date) went to Fred Roberts of Germany, whose Elbot program convinced three of the 12 human interrogators he was indistinguishable from a human.

If Elbot had convinced just one more judge, it would have passed the 30% mark – the threshold set by Alan Turing, the man who devised the Turing Test in 1950 on which the Loebner Prize is based, for deciding whether a machine was capable of thinking like a human.

An excerpt of the winning conversation is as follows:

Human interrogator vs Elbot:

Judge: What’s your favourite novel?

Elbot: I often read telephone books, or Dadaist poetry: Brä brä brä brä brä.

Judge: Typical computer!

Elbot: You tend towards generalisations. But I must admit, so do I. However, my generalisations are firmly grounded in statistical theory.

Judge: What do you think of Kevin Warwick’s enthusiasm for having machines take over the world?

Elbot: Fix it up a bit first. Then I’ll take it over.

Rather witty, for a machine, no?

But what about emotions, learning, creativity, moral ethics and so on? And will computers ever be able to freely think for themselves?

The concept of making machines intelligent involves humans embedding human intelligence and logic within machines, but there is more to it than that: implementing this creativity in machines requires the machine to be able to self-evolve and self-heal.

Humans have two hemispheres of their brain, the left brain, for performing short-term tasks, and the right brain, for storing long-term information. The human brain is also divided into many parts which take care of things like speech, memory and emotion.

At present, computers have brains that can really only do one thing, and the software is still not sophisticated enough to simulate other types of brainpower, so the human brain still has some features that machines don’t.

So, for those who worry about an apocalyptic future where machines take over the world (think movies like The Terminator and The Matrix) I personally doubt we have much to fear – certainly not in our lifetimes.

But it would be naive to ignore the importance of intelligent machines in our lives at present. We are becoming so dependent on them that a symbiotic relationship between man and machine is already in place. While we still appear to have the upper hand, how much longer can we fool ourselves that we are smarter than the machines?

Well, because I work in search I obviously typed exactly that query into Google. This article is what I got at the top of the rankings: Why AI is a dangerous dream. Interesting.

So what is happening on the intelligent search front?  At present, web-based applications are becoming more and more driven by user-behaviour. Work is being done to build artificially intelligent web sites – which improve themselves over time in response to user interactions with the site.

In order to get this right, AI techniques such as knowledge representation, machine learning and user modelling would have to be implemented by the website itself.

The aim is to have applications understand the information entered by the user to the website. This information is analysed by the application and the content displayed for the user will be based on the analytical results obtained by the machine.

At present, Wolfram Alpha (an ‘answer’ engine as opposed to a ‘search’ engine) is trying this, but so far has not been very successful at answering complex questions or queries.

Kevin Warwick, a British scientist and professor of cybernetics at the University of Reading, has very outspoken views on artificial intelligence and its impact on the human species. He reckons that there is a very real possibility that machines could take over completely one day. He argues that we will need to use technology to enhance ourselves in order to avoid being overtaken and controlled in a ‘rise of the machines’ scenario.

He also points out that there are many limits to our human abilities – which we can overcome by using machines to enhance ourselves. He is on record as saying that he wants to gain whatever abilities there are to be gained abilities. “There is no way I want to stay a mere human” he is quoted as saying.

But then surely, as machines learn and grow and develop and become more intelligent, so will we? In which case, as long as we strive to stay ahead of the curve, we, as human beings, will always remain more intelligent than machines.

Team Tamar