Core Memory  cover image

He Thinks AI Code May Break Everything - EP 59 Will Wilson

Core Memory

00:00

How AI Models Are Trained to Write Code

Will explains pre-training versus reinforcement learning and why RL with human feedback can favor undetectable bugs.

Play episode from 40:22
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app