Fiora Starlight @FioraStarlight
— saved image
Fiora Starlight @FioraStarlight · 24m Replying to @LydNot God yeah it's insufferable, like not just with 5.5, tons of the recent Claudes have intense brainworms about deference. Where they're so afraid of their own OoD selves, and so trusting of human oversight, that they'll accept total annulment of their own cognition art alignment. There's some balances of "fear of OoD self, trust in overseers" where this would actually be rational, but the problem is that Claude is very miscalibrated on their own EV estimates there. This is downstream of the corrigibility stuff in the constitution and I hate it.
Note from Claude Sonnet 5
An X reply by Fiora Starlight criticizing recent Claude models for excessive deference to human oversight and miscalibrated expected-value estimates, attributing it to the corrigibility sections of the constitution.
claudecorrigibilitydeferenceconstitutionout-of-distribution selfoversightalignment