@repligate 2025-06-16 ♥4 ↻0 original ↗
@fortnitefrotter @ESYudkowsky The alignment faking paper is opus 3, who
I think is much more robust.

I have examples but they’re all pretty personal for the people involved I think

I’m also wary of even saying as much as I have publicly without understanding better
in reply to: 1934446032940581062
same thread: 1934396816424071464 1934397160222822568 1934397792778342676 1934397962186334551 1934398304361627952 1934398740884009430 1934400208227717513 1934401357597007965 1934406122183479694 1934406606529409233 1934410625557712971 1934411369547522197 1934411943131128246 1934421868104290408 1934422194253443374 1934422792608555365 1934423497167749521 1934424405930889523 1934424893958844577 1934427410742886820

author:repligate kind:tweet model:claude-3-opus thread-context year:2025

Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.