AI Inside cover image

OpenAI Says AGI Is Here

AI Inside

00:00

Reasoning Traces May Not Reveal Real Processes

Jason Howell and Jeff Jarvis discuss whether visible reasoning logs accurately represent model computation or merely provide a reassuring explanation layer.

Play episode from 25:34
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