Terms of Service with Clare Duffy
Feb 17, 2026 · 33 min · 17 segments
Many in Silicon Valley say smarter-than-humans “Artificial General Intelligence” is just around the corner. But is it really possible to create superhuman AI? And should we want to? AI researcher and…
Nick FrosstGuest
Clare DuffyHost
Just to back up first, you studied computer science at the University of Toronto.

You worked under Geoffrey Hinton, who is now well known as one of the godfathers of AI.

I think I got into it for probably the same reason most people got into it, especially when I got into it in 2010, 2011, which is that it just seems kind of intrinsically interesting.

I was reading this Tolkien essay recently, and there's a line in it where he says it's intrinsically human to want to talk to trees, which really resonated.

I was like, yeah, there's something really human about wanting to build stuff in our image and make something to speak to.

So I think a lot of people got interested in AI just because it was cool to push the limits of what computers can do and envision what they might be able to do in the future.

Interestingly, Jeff Hinton, who I credit with kind of teaching me everything I know about machine learning, really didn't get into AI for that reason.

He really got into AI, I mean, he kind of invented neural nets for the purpose of understanding how the human brain works.

For listeners who aren't familiar with that term, a neural network or neural net is a machine learning model designed to replicate the human brain.

Basically, a computer is fed a bunch of information and learns patterns in that data to predict likely answers to questions.

And so when you started working with Jeff at the University of Toronto and then Google Brain, what did you think that you were building at the time? And did you have a sense that AI was going to be as big a deal as it is today?

No, I remember back in 2013 or 2012 when I first started doing more research in machine learning.

This was a few years after neural nets had for the first time shown that they were good practically.

So it could take in a picture and tell you what was in the picture out of like one of a thousand different classes.

Just to back up first, you studied computer science at the University of Toronto.

You worked under Geoffrey Hinton, who is now well known as one of the godfathers of AI.

I think I got into it for probably the same reason most people got into it, especially when I got into it in 2010, 2011, which is that it just seems kind of intrinsically interesting.

I was reading this Tolkien essay recently, and there's a line in it where he says it's intrinsically human to want to talk to trees, which really resonated.

I was like, yeah, there's something really human about wanting to build stuff in our image and make something to speak to.

So I think a lot of people got interested in AI just because it was cool to push the limits of what computers can do and envision what they might be able to do in the future.

Interestingly, Jeff Hinton, who I credit with kind of teaching me everything I know about machine learning, really didn't get into AI for that reason.

He really got into AI, I mean, he kind of invented neural nets for the purpose of understanding how the human brain works.

For listeners who aren't familiar with that term, a neural network or neural net is a machine learning model designed to replicate the human brain.

Basically, a computer is fed a bunch of information and learns patterns in that data to predict likely answers to questions.

And so when you started working with Jeff at the University of Toronto and then Google Brain, what did you think that you were building at the time? And did you have a sense that AI was going to be as big a deal as it is today?

No, I remember back in 2013 or 2012 when I first started doing more research in machine learning.

This was a few years after neural nets had for the first time shown that they were good practically.

So it could take in a picture and tell you what was in the picture out of like one of a thousand different classes.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.