Press Space to continue
This creates a structural problem where meaningful safety
Press Space to continue
An outside AI safety evaluator can’t assess what a lab won’t show them. Embedding evaluators means little without meaningful access, Apollo Research’s CEO argues. Read more: https://thein.fo/3VtiQ8c
open on x · 3 viewsViews and clicks count once per visitor per day. Clicks are opens of this post on X.