Hello, I'm Professor Hannah Fry. I am a mathematician who writes about artificial intelligence and its role in society.

I'm Dr Adam Rutherford and I am AI Curious.

Now, we're taking a break from our usual curious cases for a different kind of investigation.

Over the next four episodes, we're going to get stuck into the world of artificial intelligence of sentient machines.

And what all of this means for the future of humanity.

Hey, no big deal, guys. We aren't going to be wading through the weeds of some of the biggest ideas in machine learning,

delving into the depths of what data can and can't tell us about ourselves and unpicking some of the mysteries of what lies ahead.

Now, in this series, we're bouncing off the back of this year's re-flectures where the supremos of public intellectuals opine on the issues do sure.

This year's lectures are called Living with Artificial Intelligence and they're being given by Stuart Russell, Professor of Computer Science at the University of California at Berkeley,

holder of the Smith Zuddy Chair in Engineering, Director of the Centre for Human Compatible AI and author of Artificial Intelligence, a modern approach.

Seems like a appropriate list of credentials for such a role.

Now, for each of the four lectures, we're going to pick up on the theme, get hold of some of the ideas that Stuart has discussed and take them apart,

examine their circuitry and solder them back together for you. In this first program, we are starting with a surprisingly difficult question to answer.

Just what is artificial intelligence? Here's what our re-flecture Stuart Russell said on the matter.

From the very beginnings of AI, intelligence and machines has been defined in the same way.

Machines are intelligent to the extent that their actions can be expected to achieve their objectives.

But because machines, unlike humans, have no objectives of their own, we give them objectives to achieve. In other words, we build objective-achieving machines.

This whole idea started way back in the 1840s. Adelauvle said that machines would eventually be able to do whatever we know how to order them to perform.

So basically machines that could follow a recipe?

Exactly. And then Alan Cheering a full hundred years later picked up the button and added that he thought machines could give rise to

surprises. That if you set an appropriate objective, it was possible for computers to achieve it by doing something that could be described as thinking.

Right. Well, that's not following a recipe anymore. That sounds more like the sort of technical challenge on Bake-Off.

...