The Rise of AI

What is the big deal with artificial intelligence?

Back in 1997, a computer was able to defeat a human being at a game of chess. The victor, a machine called Deep Blue, managed to beat Garry Kasparov, one of the best human players of all time. The chess-playing computer was developed by IBM, written in C (a computer programming language), and thus contained enough computing power to evaluate 200 million positions per second. It used a technique known as brute-force to evaluate these millions of positions and to choose the best moves to defeat Mr. Kasparov.

But fast-forward to March of 2016, AlphaGo, a computer program developed by researchers at the London-based company DeepMind, which is now owned by Google, won a five-game match of Go against Lee Sedol, considered the game’s best player. Developing computers which can successfully defeat human opponents at this ancient East Asian game is a much more complex task than one which can play and beat a human at chess. The estimated number of possible positions of stones on a Go board is 10170.

This shows just how much computers have managed to improve over the past 20 years or so. It is not just the technologists computer scientists who are excited; economists, philosophers, politicians, lawyers and many others will take great interest in the latest developments taking place in AI (artificial intelligence), a technology allowing computers to do tasks which could traditionally only be done by humans. In addition, there are other technologies that are helping push these significant shifts. Machine learning, a way of achieving artificial intelligence, is the practice of analysing data to make determinations or predictions, essentially ‘training’ computers to complete specified tasks. Deep learning is a way to implement machine learning which has now been made possible, thanks to improved computational power and the availability of big data.

Though with the rise of AI, opinions are split. There are some experts who predict that it will have noticeable impacts sooner rather than later, whereas others believe it will still take some time before AI begins to have truly drastic impacts on the world. The impact it may have on employment is an example of where there appears to be a divide, particularly about whether future improvements in AI will leave masses of people unemployed as more advanced machines start to become the preferred choice of capital.

NEXT |The Big Questions

The challenges that the rise of AI could bring