Google DeepMind: The Podcast · AI Research & Frontier Labs · November 2026
Fry asked whether the grind of building commercial agents (flights, restaurants, calendars) pulls effort away from advancing intelligence. He says no: a general intelligence that can't book a restaurant is, by definition, spiky.
that is by making the models part of our life.
…do. You have issues with cybersecurity, with agent protocols. I mean, there's a lot of faff, basically, in getting this stuff to actually work. So, I guess my question is: when you're building these commercial products, does that friction end up pulling you away from advancing intelligence itself?
Not at all. Like, I mean, like, because if we don't, then people get to ask the question: you can do all these things, but you can't book a restaurant for me. And that is the definition of spikiness. And if we are talking about a general intelligence, like if I can solve all these hard problems, I got to be able to book a restaurant.
Surely.
Surely. So that's part of it. I wouldn't make too much of a differentiation between, oh, these kinds of tasks are important. That is why let's define intelligence as that. And these kinds of tasks are not important. And that's the importance about products. When you do those, you have to solve…
I really like that idea. But I also, it does make me wonder. I wonder about whether you're giving yourselves a sort of bigger challenge than anyone else because you're trying to integrate Gemini into maps and documents and email and calendar and so on, which have like billions of users each. That's…