Jonathan McRae

I don't know how many studies we can do on it, but what was this about?

Someone on the internet with a crazy team.

We're spending money on research where the evidence base is not very high, but you have people in government, particularly in the United States, who are, you know, who are giving credence to, I don't know, like, I mean, that doesn't seem like an enormous amount of evidence to start a public.

I mean, conversations are really important.

I suppose just saying, just trust me on this hasn't worked very well in the past.

So you're absolutely right, Matthew.

Matthew Campbell and Jasmine Fairfield, thank you very much.

Thank you.

All right, on the way, are scientists soon to be a thing of the past?

Welcome back.

Now, in the coming years, I'm sure there'll be a lot of people out there who might be inclined to curse computer scientists for developing AI to the point that it has made their job obsolete.

But if it makes you feel any better, it appears they're intent on doing away with their own profession as well.

Jeff Klune is CIFAR AI chair at the Victor Institute and professor of computer science at the University of British Columbia in Vancouver.

He's at the forefront of developing an AI scientist.

He joins me now.

Jeff, how exactly does a scientist AI work?

Okay, so the idea of that was ludicrous a few years ago, but as these AI systems have become increasingly better at thinking and performing cognitive tasks, this is no longer an idea.

This is actually a reality.

You've actually built something that has been able to produce a paper that's been published in a scientific journal?

So this is, of course, limited to a few domains, right?