The Inside View cover image

3. Evan Hubinger on Takeoff speeds, Risks from learned optimization & Interpretability

The Inside View

00:00

Intent Enlightenment

I'm a cross sort of different possible situations. Intent linmet is like trying to not do bad things. So intensilimet can be split into am outer alignment, which is the objective that is trained on alind and then this objective, robusinous thing, which is because it is actually robust according to that objective. Ye, i was trying to find the actual actual picture from the blood cote because free good a but y having a sen clarifying in arloan terminology.

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