The Meaning Code cover image

Computation is the Covenant Between Truth and Language

The Meaning Code

00:00

Aligning AI to Human

The way to hack large language models is to flood the learning set with false statements. In a sense they are running a program the are intelligent, you know, in a loose sense. They don't have the ability to step outside and say no, that doesn't work. But it then gets into the whole alignment question. Is that a question of language alignment alignment aligning AI to human?

Transcript
Play full episode

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