Predicting AI's future

In a special BBC News Online webcast from the Massachusetts Institute of Technology, reporter Kevin Anderson spoke to three leading AI experts: Ray Kurzweil, author of two books on AI; Dr Rodney Brooks, the director of MIT's artificial intelligence laboratory; and Colin Angle, CEO of the company iRobot.

---

 Click here to watch the webcast.

---

Predicting the future is always a hit and miss proposition, writes Kevin Anderson.

In the 1940s, Thomas Watson, the head of IBM, famously predicted the world demand for computers might be as high as five.

And artificial intelligence has had its share of off-target predictions. AI researchers in the 1950s predicted that a computer would be the world chess champion by 1968. It took a few more decades than that.

But AI experts remain optimistic and are predicting that by the middle of the century, intelligent machines will be all around us.

AI is everywhere

In fact, they point out that artificial intelligence already pervades our lives.

Machines will gradually become more intelligent and become more pervasive

Fuel injection systems in our cars use learning algorithms. Jet turbines are designed using genetic algorithms, which are both examples of AI, says Dr Rodney Brooks, the director of MIT's artificial intelligence laboratory.

Every cell phone call and e-mail is routed using artificial intelligence, says Ray Kurzweil, an AI entrepreneur and the author of two books on the subject, The Age of Intelligent Machines and The Age of Spiritual Machines.

"We have hundreds of examples of what I call narrow AI, which is behaviour that used to require an intelligent adult but that can now be done by a computer," Mr Kurzweil says.

"It is narrow because it is within a specific domain, but the actual narrowness is gradually getting a bit broader," he adds.

The near future

Right now, Dr Brooks says that artificial intelligence is about at the same place the personal computer industry was in 1978.

In 1978, the Apple II was a year old and Atari had just introduced the 400 and 800. The choice of personal computers was pretty limited and what they could do was also relatively limited by today's standards.

Who would have thought by 2001, you would have four computers in your kitchen?

Dr Rodney Brooks, director MIT AI Lab

The metaphor may undersell AI's successes. AI already is used in pretty advanced applications including helping with flight scheduling or reading X-rays.

But the popular conception of AI as seen with HAL in 2001, Commander Data in Star Trek, and David in the film AI, is not far away, Mr Kurzweil says.

Within 30 years, he believes that we will have an understanding of how the human brain works that will give us "templates of intelligence" for developing strong AI.

And Dr Brooks says that by 2050, our lives will be populated with all kinds of intelligent robots.

Sounds outlandish? "Who would have thought by 2001, you would have four computers in your kitchen," he says, pointing to the computer chips in our coffee makers, refrigerators, stoves and radios.

Gradual change

But will our hyper-intelligent coffee makers in 2050 suddenly decide to kill us like HAL in 2001? Or will humans be made redundant by a legion of intelligent machines?

A scientist does not just wake up and decide to build a robot with emotions, Dr Brook says

No. Firstly, Dr Brooks and Mr Kurzweil believe that we will not wake up one day to find our lives populated with all manner of artificially intelligent devices.

Referring to Spielberg's movie AI in which a company creates a robot that bonds emotionally like a child, Dr Brooks says: "A scientist doesn't wake up one day and decide to make a robot with emotions."

Despite the rapid advance of technology, the advent of strong AI will be a gradual process, they say.

"The road from here to there is through thousands of these benign steps," Mr Kurzweil says.