Login
From:
www.lesswrong.com
(Uncensored)
subscribe
Empirical Observations of Objective Robustness Failures — LessWrong
https://www.lesswrong.com/posts/iJDmL7HJtN5CYKReM/empirical-observations-of-objective-robustness-failures
links
backlinks
Roast topics
Find topics
Find it!
Inner alignment and objective robustness have been frequently discussed in the alignment community since the publication of “Risks from Learned Optim…