Why Does It Sometimes Pay to Be a Chicken? With Professor Michael Wooldridge

**Mia Sarenti** (0:00)
Welcome to Intelligence Squared, Where Great Minds Meet. I'm producer Mia Sarenti.
Why does it sometimes make sense to swerve, compromise or cooperate, and at other times to stand your ground? From international crises to everyday decisions, how can we think more strategically about the choices we make? On today's episode, Professor Michael Wooldridge, Asshol Professor of the Foundations of Artificial Intelligence at the University of Oxford, sheds light on how game theory, one of mathematics' most influential fields, can offer some surprising life lessons when it comes to conflict, cooperation and human behaviour. Let's join our host, Carl Miller, now with more.

**Carl Miller** (0:39)
Welcome to Intelligence Squared, everyone. I'm Carl Miller and our returning guest today is Professor Mike Wooldridge.
Michael Wooldridge is the actual Professor of the Foundations of Artificial Intelligence in the Department of Computer Science at the University of Oxford, a Senior Research Fellow at Hertford College. He's been an AI researcher since 1989, before AI was cool, Mike, and has published more than 400 scientific articles on the subject. His latest book is Life Lessons from Game Theory, the Art of Thinking Strategically in a Complex World. Michael, you've been on Intelligence Squared before. A very warm welcome back to you.

**Michael Wooldridge** (1:16)
Great to be back.

**Carl Miller** (1:17)
So I'm a firm believer in beginning at the beginning when it comes to these sorts of subjects, Michael. So the first question might be a very simple one, but can you just explain what game theory actually is for people that might not have ever heard of it before, or of course used it?

**Michael Wooldridge** (1:34)
So game theory is the mathematical theory that tries to understand every situation in which self-interested parties interact. And those parties could be me and you, they could be governments, they could be nations engaged in a bloody war that are trying to navigate their way through it. They could be anything which is perceived of as having choices that can take actions and also has self-interest in the sense that it has preferences about how the world is. So game theory was actually invented as far back as the 1920s.
And one of the main inventors was the Hungarian-American mathematician John von Neumann. And von Neumann was a mathematician who was interested in games. And he asked himself a very, very, very natural question for a mathematician. He said, could game theory teach me how to play poker well? Now, von Neumann solved that problem mathematically. And in doing so, he kind of set off in train the whole of the field of game theory. He didn't actually tell us how to play poker well, but he established the main theory that was underpinning it. The problem is, when people hear game theory as a consequence of that, they think it's only to do with parlor games, games like chess or poker or checkers and so on. But actually, game theory studies every scenario where self-interested parties interact with one another. And that could be people playing a game of poker and in a game of poker, it's obvious what their self-interest is. They want to win. If I'm playing with you, I want to win the game. You also want to win the game. So our preferences there are not aligned with one another. And so the theory tells you how you might behave in such a game.
The players could be, it could be me and my wife negotiating about who's going to wash up the dishes and who gets to choose what we watch on TV tonight. They could be a parent negotiating with a child about how many screen time, how much screen time they get before they go to bed. Could be you and your boss negotiating about your pay rise or two companies trying to reach a deal in the market. So although it has its origins in parlour games, the scope of game theory is very, very, very broad indeed.

**Carl Miller** (3:53)
And so if we just stay with a poker example and Vaughan Oiman just for one more moment, what does that look like? So if one is applying kind of mathematical principles of kind of analysis or conjecture to a problem like poker, what actually are you doing? How does math help you think about decision making in that particular game?

**Michael Wooldridge** (4:14)
So there is a big branch of mathematics called optimisation, which is about trying to compute the optimal choice given a range of possibilities and your preferences over this. And it's a very big, very, very well established theory.
And what game theory does is try to say, how do we do that in a game of poker? What's the choice that I could make, the move I could make in this game, which would result in the best outcome for me? Now what Von Neumann did is, I mean, he had to greatly simplify things because poker is a complex game, but he came up with a simple model of what are called zero sum games. And a zero sum game is basically a win-lose game. And a game of poker is a zero sum game in some sense, because if I win, you lose, right?