Revolution.Social cover image

Jeff Jarvis on the Death of Mass Media, Twitter vs. UberMedia, and Section 230’s Brilliance

Revolution.Social

00:00

Bias in training data and LLM behavior

Evan notes training data sources (like Reddit) shape LLM values, so models reflect dataset biases and choices.

Play episode from 46:31
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