Gods and Robots

Katherine Gorman (0:07)

You are listening to Talking Machines. I'm Katherine Gorman.

Michael Littman (0:10)

And I'm Michael Littman.

Katherine Gorman (0:11)

Michael, thank you so much for joining me on the host side of the table today. I know it's a little unusual, but it's because we've got Neil on the guest side of things.

And our listeners might know you from your appearances as a guest on the show multiple times. But before we dive in, tell us a little bit about yourself, Michael.

Michael Littman (0:28)

Oh, thanks, Katherine. Yeah, I'm a machine learning researcher and a professor at Brown University. I mainly work in the reinforcement learning corner of the field. I've gotten more and more interested lately in exploring reinforcement learning as a powerful tool for people to teach machines what we want them to do.

I've also been trying to help out however I can with communicating about machine learning. You and I work as the NeurIPS communication co-chairs together, and I'm really glad you asked me to be on the podcast today.

Katherine Gorman (0:54)

Oh, I'm so glad that you said yes, Michael. I hope it's not too painful.

Michael and I are going to bring you a very special episode focusing on an amazing project that Neil is involved with. And so, of course, we have Neil with us today. Neil, tell us about yourself as a guest.

Neil Lawrence (1:10)

Oh, well, I'm a professor of machine learning at the University of Cambridge, and normally co-host of Talking Machines podcasts, but not today.

Katherine Gorman (1:18)

Not today, Neil, not today. And also with us, we have this amazing collaborator, Rabbi Laura Janner-Klausner. Rabbi, welcome. Thank you so much for taking the time to talk with us.

Laura Janner-Klausner (1:28)

Oh, it's wonderful to be here, really.

Katherine Gorman (1:30)

Fantastic.

So we usually ask all of our guests the same question first. How did you get where you are? But I think we have a pretty good scope on Neil. But Rabbi, tell us about your path. How did you come to the place where you are now?

Laura Janner-Klausner (1:44)

England, London, Jerusalem, London. Teaching, Palestinian-Israeli dialogue facilitation, youth leaders teaching.

Then came back to London and studied for the Rabbinate, which is five years, and then became a Rabbi. And then I was the senior Rabbi for nearly a decade of reformed Judaism in the UK. And one of the things that happened through that is I met Neil, and I am a geek, love AI and technology. And so the discussions around ethics are core because I believe that as a Rabbi, well really any religious leader needs to get their hands into what are the kind of the ethical issues of the day. And this is writ large capital letters, ethical issue of the day.

Michael Littman (2:32)

Neil, we're pretty well versed in how you got to where you are academically, but tell us about how you met Rabbi Lor.

Neil Lawrence (2:38)

Actually, I guess it goes back to around the time when machine learning, artificial intelligence was sort of hitting the news. I read the Superintelligence Book by Nick Bostrom and I was reading it because I felt I had to read it to, I disagree with a lot of what's in it, but I had to read it to understand what he was saying. And I was really, really struck by the parallels between what's written about in that book and what a lot of people were saying about that book and what I sometimes think of as cartoon religion. So the notion of a sort of omniscient, omnipotent intelligence that has the power to control us all and that we should respond by perhaps trying to seek immortality ourselves.

And I thought, wow, it almost appears like what we're creating here is sort of religion by the back door through technology. And it started me thinking about relationship between what's going on. I'm not religious myself, well, I'm not a believer, but I just started seeing a lot of parallels between what religious groups talk about. And I think also in this very cartoon way, I think a very simplistic way, what people are talking about sort of in the super intelligence argument. But I think more interestingly, there's a deeper connection in the way that religious people think about the context of being human in the context of what they see as a sort of existing higher power or a God. And I think that that somehow actually causes them to introspect a lot about what human intelligence is.

And I really like the idea that artificial intelligence will help us introspect a lot more about human intelligence.

Michael Littman (4:25)