3. Evan Hubinger on Takeoff speeds, Risks from learned optimization & Interpretability
The Inside View
00:00
Amplified Overseader and Interpretability Space
i have some belief that we can probably just avoid the drop off at the end if we use an amplified overseer to do transparency. Ote said fagtously. And an oversiht overseer. What he said? Amplified overs oh, emptified overseer ys. Both of us are trying to do some field building in the linment and crystallize,. maybe more to about a fieldbuilding in the interpretability space.
Transcript
Play full episode
