WUOMFM

Computer beats human in Go: Should we be scared or excited about A.I. milestone?

Mar 9, 2016

The AlphaGo computer beating a human champion in the game of Go likely won't lead to this. Or will it?
Credit User: Insomnia Cured Here / Flickr - http://j.mp/1SPGCl0

For decades, science fiction movies and books have predicted that someday machines will develop artificial intelligence (A.I.) and take over the world. While it’s not exactly the plot of a prequel in the Terminator movie franchise, world champion Lee Sedol was defeated by a computer in a game of Go this week. This marked a milestone for the development of A.I.

Google’s DeepMind program, named AlphaGo, was victorious over the strategy game’s best player of the last decade (watch the video of the match below). It was the first game in a best-of-five series between the two with $1 million in prize money on the line.

To help us understand more about what exactly artificial intelligence is, Michigan State University biologist and computer scientist Arend Hintze joined Stateside to talk about his research in the field.

Hintze has said that he believes that artificial intelligence is not just for sci-fi movies, but rather something that could become a reality in his lifetime.

I'm not too sure that we don't have artificial intelligence right now, it's just not very smart yet.

“The founding fathers of A.I., like Alan Turing and those people, thought about this really deeply, what is intelligence and what is artificial intelligence,” said Hintze. “They came up with something called the Turing Test. The Turing Test means that you talk to the A.I. and you make up your mind about this being a program or this being human-like or as much of a human as it can be.”

We’ve seen computers defeat humans in games like chess and Jeopardy, but Hintze says those are single-purpose machines, and what he is trying to accomplish is creating a digital brain that evolves like the human brain.  

Watch AlphaGo face off with Lee Sedol in a game of Go:

Just how far away are we from A.I. being a part of our day-to-day lives?

“I’m not too sure that we don’t have artificial intelligence right now; it’s just not very smart yet,” said Hintze. “So the real question is, if we get human-level intelligence or something that is at least comparable in certain degrees. It is really limited on two questions. The first thing is do we have enough computational resources? And can we comprehend evolutionary mechanics faster than we can actually understand how brains work from a neuroscience point of view? The other sciences have tried to understand human-level intelligence or behavior and technically we wouldn’t be able to reverse-engineer this until we understand how this is working."

Listen to the full interview below to hear more about the developments and goals of A.I. and Hintze’s interpretation of the computer HAL in the movie 2001: A Space Odyssey.