80,000 Hours Podcast cover image

Why Silicon Valley can’t slow down AI on its own | Luisa Rodriguez

80,000 Hours Podcast

00:00

Safety Checkpoints For Dangerous Capabilities

Luisa Rodriguez examines capability-based checkpoints, their unclear thresholds, and the difficulty of proving models are safe when evaluations can be deceived.

Play episode from 06:06
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