RS Rich Sutton Argues that agents unable to keep learning from experience after deployment are capped from the start, a view described as the big world hypothesis: the world i...

“I'm not weird. The field is weird. The field, they need to call it continual learning. It's just learning.”

Training Data · AI Research & Frontier Labs · August 2026

“I'm not weird. The field is weird. The field, they need to call it continual learning. It's just learning.” — Rich Sutton, Training Data

Context

He opens the episode with it and returns to it word for word an hour later. People keep telling Sutton his position is radical; he thinks the rest of the field started thinking strangely about a decade ago — before the AI boom, nobody needed the phrase "continual learning" because no other kind existed.

Transcript

Training Data Around 00:00 into the episode
Rich Sutton

People think I have a radical point of view sometimes. They start questions saying how what I'm thinking is so different from everyone else. But I don't see it that way at all. I see it as like I'm thinking the ordinary way. It's just everyone else that's thinking a bit weird. And I mean that, like, you know, it's just the recent times people are thinking weird. Before there was all this AI craziness, you talk about, you wouldn't have to say continual learning because it wouldn't make any sense to talk about learning that wasn't continual. All learning is continual. We always act and we learn. That's just the normal way of thinking. I'm not weird. The field is weird. The field, they need to call it continual learning. It's just learning.

Sonya Huang

We are honored to have the great Rich Sutton with us here today. Rich, you invented reinforcement learning. You wrote the seminal textbook. You're the key students in the field, folks like Dave Silver. You wrote the essay, The Bitter Lesson, that I believe is the Bible of the field. And you have…

Rich Sutton

What else are you going to

Sonya Huang

do?

Speaker names come from our own diarization. The position is estimated from where the line sits in the episode.

More from Training Data