Eye On A.I. cover image

#123 Aidan Gomez: How AI Language Models Will Shape The Future

Eye On A.I.

00:00

How to Deal With Hallucinations

You have to distinguish between hallucinations that you want, which are like creativity and the hallucinations that you don't want. Another really good way to get models to be more truthful is to actually force them to cite their work. If I mess up, if the user clicks through and sees a, an error or a hallucination, I'm going to get a super strong negative feedback.

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