How to Eliminate Bias From Neural Network Learning Systems

The AIs can and do actually multiple levels of reasoning where they check their own initial responses and then iterate based on it before they send a response. But again, they're not being explicitly trained on this. No one is telling them, here's the taxonomy of all evolution of felines. It's just kind of figuring that out on its own. That's actually the black box that I worry about more.

Transcript

Play full episode

Transcript

Episode notes

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app